Вы находитесь на странице: 1из 666

# Numerical Analysis for

## Scientists and Engineers:

Theory and C Programs

Department of Applied Mathematics with
Oceanology and Computer Programming
Vidyasagar University
Midnapore - 721102

Dedicated to my parents

Preface
Numerical Analysis is a multidisciplinary subject. It has formed an integral part of
Physics, Commerce and dierent Engineering streams. Numerical Analysis shows the
way to obtain numerical answers to applied problems. Numerical methods stand there
where analytical methods may fail or complicated to solve the problem. For example, in
nding the roots of transcendental equations or in solving non-linear dierential equations. So, it is quite impossible to train the students in applied sciences or engineering
without an adequate knowledge of numerical methods.
The book is suitable for undergraduate as well as for postgraduate students and
advanced readers. Ample material is presented so that instructors will able to select
topics appropriate to their needs. The book contains ten chapters.
In Chapter 1, dierent types of errors and their sources in numerical computation
are presented. The representation of oating point numbers and their arithmetic are
studied in this chapter.
Finite dierence operators, relations among them are well studied in Chapter 2. The
dierence equations and their solution methods are also introduced here.
Chapter 3 is devoted to single and bivariate interpolations. Dierent types of interpolation methods such as Lagrange, Newton, Bessal, Stirling, Hermite, Everette are
incorporated here. Inverse and cubic spline interpolation techniques are also presented
in this chapter. Several bivariate methods are presented here.
Several methods such as graphical, tabulation, bisection, regula-falsi, xed point
iteration, Newton-Raphson, Aitken, secant, Chebyshev and Muller are well studied to
solve an algebraic and transcendental equation in Chapter 4. The geometrical meaning
and the rate of convergence of the above methods are also presented. The very new
method, modied Newton-Raphson with cubic convergence is introduced here. BirgeVieta, Bairstow and Graees root squaring methods are deduced and illustrated to
nd the roots of a polynomial equation. The methods to solve a system of non-linear
equations are introduced here.
Chapter 5 deals to solve a system of linear equations. Dierent direct and iterative
methods such as matrix inverse, Gauss-Jordon, Gauss elimination, LU decomposition,
vii

## viii Numerical Analysis

Cholesky, matrix partition, Jacobi, Gauss-Seidal and relaxation are well studied here.
The very new methods to nd tri-diagonal determinant and to solve a tri-diagonal system of equations are incorporated. A method to solve ill-condition system is discussed.
The generalised inverse of a matrix is introduced. Also, least squares solution method
for an inconsistent system is illustrated here.
Determination of eigenvalues and eigenvectors of a matrix is very important problem
in applied science and engineering. In Chapter 6, dierent methods, viz., LeverrierFaddeev, Rotishauser, Power, Jacobi, Givens and Householder are presented to nd the
eigenvalues and eigenvectors for arbitrary and symmetric matrices.
Chapter 7 contains indepth presentation of several methods to nd derivative and
integration of a functions. Three types of integration methods, viz., Newton-Cotes
(trapezoidal, Simpson, Boole, Weddle), Gaussian (Gauss-Legendre, Lobatto, Radau,
Gauss-Chebyshev, Gauss-Hermite, Gauss-Leguerre, Gauss-Jacobi) and Monte Carlo are
well studies here. Euler-Maclaurin sum formula, Romberg integration are also studied
in this chapter. An introduction to nd double integration is also given here.
To solve ordinary dierential equations, Taylor series, Picard, Euler, Runge-Kutta,
Runge-Kutta-Fehlberg, Runge-Kutta-Butcher, Adams-Bashforth-Moulton, Milne, nitedierence, shooting and nite element methods are discussed in Chapter 8. Stability
analysis of some methods are also done.
An introduction to solve partial dierential equation is given in Chapter 9. The nite
dierence methods to solve parabolic, hyperbolic and elliptic PDEs are discussed here.
Least squares approximation techniques are discussed in Chapter 10. The method
to t straight line, parabolic, geometric, etc., curves are illustrated here. Orthogonal polynomials, their applications and Chebyshev approximation are discussed in this
chapter.
The algorithms and programmes in C are supplied for the most of the important
methods discussed in this book.
At rst I would like to thank Prof. N. Dutta and Prof. R.N. Jana, as from their book
I learnt my rst lessons in the subject.
In writing this book I have taken help from several books, research articles and some
websites mentioned in the bibliography. So, I acknowledge them gratefully.
This book could not have been complete without the moral and loving support and
also continuous encouragement of my wife Anita and my son Aniket.
Also, I would like to express my sincere appreciation to my teachers and colleagues
specially Prof. M. Maiti, Prof. T.K. Pal and Prof. R.N. Jana as they have taken all
me with sucient time to write this book. I would also like to acknowledge my other
colleagues Dr. K. De and Dr. S. Mondal for their encouragement.
I express my sincerest gratitude to my teacher Prof. G.P.Bhattacharjee, Indian Institute of Technology, Kharagpur, for his continuous encouragement.

ix
I feel great reverence for my parents, sisters, sister-in-law and relatives for their
blessings and being a constant source of inspiration.
I would like to thank to Sk. Md. Abu Nayeem, Dr. Amiya K. Shyamal, Dr. Anita
Saha, for scrutinizing the manuscript.
I shall feel great to receive constructive criticisms for the improvement of the book
from the experts as well as the learners.
I thank the Narosa Publishing House Pvt. Ltd. for their sincere care in the
publication of the book.

x Numerical Analysis

Contents
1 Errors in Numerical Computations
1.1 Sources of Errors . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Exact and Approximate Numbers . . . . . . . . . . . . . . .
1.3 Absolute, Relative and Percentage Errors . . . . . . . . . .
1.4 Valid Signicant Digits . . . . . . . . . . . . . . . . . . . . .
1.5 Propagation of Errors in Arithmetic Operations . . . . . . .
1.5.1 The errors in sum and dierence . . . . . . . . . . .
1.5.2 The error in product . . . . . . . . . . . . . . . . . .
1.5.3 The error in quotient . . . . . . . . . . . . . . . . . .
1.5.4 The errors in power and in root . . . . . . . . . . . .
1.5.5 Error in evaluation of a function of several variables
1.6 Signicant Error . . . . . . . . . . . . . . . . . . . . . . . .
1.7 Representation of Numbers in Computer . . . . . . . . . . .
1.8 Arithmetic of Normalized Floating Point Numbers . . . . .
1.8.1 Addition . . . . . . . . . . . . . . . . . . . . . . . . .
1.8.2 Subtraction . . . . . . . . . . . . . . . . . . . . . . .
1.8.3 Multiplication . . . . . . . . . . . . . . . . . . . . . .
1.8.4 Division . . . . . . . . . . . . . . . . . . . . . . . . .
1.9 Eect of Normalized Floating Point Representations . . . .
1.9.1 Zeros in oating point numbers . . . . . . . . . . . .
1.10 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 Calculus of Finite Di. and Di. Equs
2.1 Finite Dierence Operators . . . . . . . . . . .
2.1.1 Forward dierences . . . . . . . . . . . .
2.1.2 Backward dierences . . . . . . . . . . .
2.1.3 Central dierences . . . . . . . . . . . .
2.1.4 Shift, Average and Dierential operators
2.1.5 Factorial notation . . . . . . . . . . . .
2.2 Properties of Forward Dierences . . . . . . . .
xi

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

1
1
2
4
7
9
9
10
11
14
15
16
18
19
19
20
21
21
22
23
23

.
.
.
.
.
.
.

27
27
27
28
29
30
31
32

## xii Numerical Analysis

2.2.1 Properties of shift operators . . . . . . . . . . .
2.3 Relations Among Operators . . . . . . . . . . . . . . .
2.4 Representation of Polynomial using Factorial Notation
2.5 Dierence of a Polynomial . . . . . . . . . . . . . . . .
2.6 Summation of Series . . . . . . . . . . . . . . . . . . .
2.7 Worked out Examples . . . . . . . . . . . . . . . . . .
2.8 Dierence Equations . . . . . . . . . . . . . . . . . . .
2.8.1 Formation of dierence equations . . . . . . . .
2.9 Solution of Dierence Equations . . . . . . . . . . . .
2.9.1 Iterative method . . . . . . . . . . . . . . . . .
2.9.2 Solution using symbolic operators . . . . . . .
2.9.3 Generating function . . . . . . . . . . . . . . .
2.10 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

33
34
39
43
45
47
52
53
55
55
56
63
66

3 Interpolation
3.1 Lagranges Interpolation Polynomial . . . . . . . . . . . . . . . . .
3.1.1 Lagrangian interpolation formula for equally spaced points
3.2 Properties of Lagrangian Functions . . . . . . . . . . . . . . . . . .
3.3 Error in Interpolating Polynomial . . . . . . . . . . . . . . . . . . .
3.4 Finite Dierences . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4.1 Forward dierences . . . . . . . . . . . . . . . . . . . . . . .
3.4.2 Backward dierences . . . . . . . . . . . . . . . . . . . . . .
3.4.3 Error propagation in a dierence table . . . . . . . . . . . .
3.5 Newtons Forward Dierence Interpolation Formula . . . . . . . .
3.5.1 Error in Newtons forward formula . . . . . . . . . . . . . .
3.6 Newtons Backward Dierence Interpolation Formula . . . . . . . .
3.6.1 Error in Newtons backward interpolation formula . . . . .
3.7 Gaussian Interpolation Formulae . . . . . . . . . . . . . . . . . . .
3.7.1 Gausss forward dierence formula . . . . . . . . . . . . . .
3.7.2 Remainder in Gausss forward central dierence formula . .
3.7.3 Gausss backward dierence formula . . . . . . . . . . . . .
3.7.4 Remainder of Gausss backward central dierence formula .
3.8 Stirlings Interpolation Formula . . . . . . . . . . . . . . . . . . . .
3.9 Bessels Interpolation Formula . . . . . . . . . . . . . . . . . . . . .
3.10 Everetts Interpolation Formula . . . . . . . . . . . . . . . . . . . .
3.10.1 Relation between Bessels and Everetts formulae . . . . . .
3.11 Interpolation by Iteration (Aitkens Interpolation) . . . . . . . . .
3.12 Divided Dierences and their Properties . . . . . . . . . . . . . . .
3.12.1 Properties of divided dierences . . . . . . . . . . . . . . .
3.13 Newtons Fundamental Interpolation Formula . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

71
72
75
77
79
87
87
88
88
90
92
95
98
99
99
102
102
105
105
106
109
111
113
116
117
121

Contents

## 3.14 Deductions of other Interpolation Formulae from Newtons

Divided Dierence Formula . . . . . . . . . . . . . . . . . . . . . . .
3.14.1 Newtons forward dierence interpolation formula . . . . . .
3.14.2 Newtons backward dierence interpolation formula . . . . .
3.14.3 Lagranges interpolation formula . . . . . . . . . . . . . . . .
3.15 Equivalence of Lagranges and Newtons divided
dierence formulae . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.16 Inverse Interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.16.1 Inverse interpolation based on Lagranges formula . . . . . .
3.16.2 Method of successive approximations . . . . . . . . . . . . . .
3.16.3 Based on Newtons backward dierence interpolation formula
3.16.4 Use of inverse interpolation to nd a root of an equation . . .
3.17 Choice and use of Interpolation Formulae . . . . . . . . . . . . . . .
3.18 Hermites Interpolation Formula . . . . . . . . . . . . . . . . . . . .
3.19 Spline Interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.19.1 Cubic spline . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.20 Bivariate Interpolation . . . . . . . . . . . . . . . . . . . . . . . . . .
3.20.1 Local matching methods . . . . . . . . . . . . . . . . . . . . .
3.20.2 Global methods . . . . . . . . . . . . . . . . . . . . . . . . . .
3.21 Worked out Examples . . . . . . . . . . . . . . . . . . . . . . . . . .
3.22 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 Sol. of Algebraic and Transcendental Equs.
4.1 Location of Roots . . . . . . . . . . . . . . . . . . .
4.1.1 Graphical method . . . . . . . . . . . . . .
4.1.2 Method of tabulation . . . . . . . . . . . .
4.2 Bisection Method . . . . . . . . . . . . . . . . . . .
4.3 Regula-Falsi Method (Method of False Position) .
4.4 Iteration Method or Fixed Point Iteration . . . . .
4.4.1 Estimation of error . . . . . . . . . . . . . .
4.5 Acceleration of Convergence: Aitkens 2 -Process
4.6 Newton-Raphson Method or Method of Tangent .
4.6.1 Convergence of Newton-Raphson method .
4.7 Newton-Raphson Method for Multiple Root . . . .
4.8 Modication on Newton-Raphson Method . . . . .
4.9 Modied Newton-Raphson Method . . . . . . . . .
4.10 Secant Method . . . . . . . . . . . . . . . . . . . .
4.10.1 Convergence of secant method . . . . . . .
4.11 Chebyshev Method . . . . . . . . . . . . . . . . . .
4.12 Muller Method . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

xiii

.
.
.
.

.
.
.
.

124
124
124
125

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

126
131
132
132
134
135
136
139
141
142
156
156
159
167
181

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

189
190
190
191
193
198
202
204
208
210
213
218
219
222
226
226
230
230

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

## xiv Numerical Analysis

4.13 Roots of Polynomial Equations . . . . . . .
4.13.1 Domains of roots . . . . . . . . . . .
4.14 Birge-Vieta Method . . . . . . . . . . . . .
4.15 Bairstow Method . . . . . . . . . . . . . . .
4.16 Graees Root Squaring Method . . . . . .
4.17 Solution of Systems of Nonlinear Equations
4.17.1 The method of iteration . . . . . . .
4.17.2 Seidal method . . . . . . . . . . . .
4.17.3 Newton-Raphson method . . . . . .
4.18 Exercise . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

237
238
239
245
253
261
261
262
266
271

## 5 Solution of System of Linear Equations

5.1 Cramers Rule . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1.1 Computational aspect of Cramers rule . . . . . . . .
5.2 Evaluation of Determinant . . . . . . . . . . . . . . . . . . .
5.3 Inverse of a Matrix . . . . . . . . . . . . . . . . . . . . . . .
5.3.1 Gauss-Jordan Method . . . . . . . . . . . . . . . . .
5.4 Matrix Inverse Method . . . . . . . . . . . . . . . . . . . . .
5.5 Gauss Elimination Method . . . . . . . . . . . . . . . . . .
5.6 Gauss-Jordan Elimination Method . . . . . . . . . . . . . .
5.7 Method of Matrix Factorization . . . . . . . . . . . . . . . .
5.7.1 LU Decomposition Method . . . . . . . . . . . . . .
5.8 Gauss Elimination Method to the Find Inverse of a Matrix
5.9 Cholesky Method . . . . . . . . . . . . . . . . . . . . . . . .
5.10 Matrix Partition Method . . . . . . . . . . . . . . . . . . .
5.11 Solution of Tri-diagonal Systems . . . . . . . . . . . . . . .
5.12 Evaluation of Tri-diagonal Determinant . . . . . . . . . . .
5.13 Vector and Matrix Norms . . . . . . . . . . . . . . . . . . .
5.14 Ill-Conditioned Linear Systems . . . . . . . . . . . . . . . .
5.14.1 Method to solve ill-conditioned system . . . . . . . .
5.15 Generalized Inverse (g-inverse) . . . . . . . . . . . . . . . .
5.15.1 Grevilles algorithm for Moore-Penrose inverse . . .
5.16 Least Squares Solution for Inconsistent Systems . . . . . . .
5.17 Jacobis Iteration Method . . . . . . . . . . . . . . . . . . .
5.17.1 Convergence of Gauss-Jacobis iteration . . . . . . .
5.18 Gauss-Seidals Iteration Method . . . . . . . . . . . . . . .
5.18.1 Convergence of Gauss-Seidals method . . . . . . . .
5.19 The Relaxation Method . . . . . . . . . . . . . . . . . . . .
5.20 Successive Overrelaxation (S.O.R.) Method . . . . . . . . .
5.21 Comparison of Direct and Iterative Methods . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

275
276
277
278
287
287
293
297
302
304
304
313
314
317
320
325
326
327
329
330
331
334
338
339
344
346
352
354
358

Contents

xv

## 5.22 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

6 Eigenvalues and Eigenvectors of a Matrix
6.1 Eigenvalue of a Matrix . . . . . . . . . . . . . . . . . . . . . . . .
6.2 Leverrier-Faddeev Method to Construct Characteristic Equation
6.2.1 Eigenvectors using Leverrier-Faddeev method . . . . . . .
6.3 Eigenvalues for Arbitrary Matrices . . . . . . . . . . . . . . . . .
6.3.1 Rutishauser method . . . . . . . . . . . . . . . . . . . . .
6.3.2 Power method . . . . . . . . . . . . . . . . . . . . . . . .
6.3.3 Power method for least eigenvalue . . . . . . . . . . . . .
6.4 Eigenvalues for Symmetric Matrices . . . . . . . . . . . . . . . .
6.4.1 Jacobis method . . . . . . . . . . . . . . . . . . . . . . .
6.4.2 Eigenvalues of a Symmetric Tri-diagonal Matrix . . . . .
6.4.3 Givens method . . . . . . . . . . . . . . . . . . . . . . . .
6.4.4 Householders method . . . . . . . . . . . . . . . . . . . .
6.5 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7 Dierentiation and Integration
7.1 Dierentiation . . . . . . . . . . . . . . . . . . . . . . . . . .
7.1.1 Error in Numerical Dierentiation . . . . . . . . . . .
7.2 Dierentiation Based on Newtons Forward Interpolation
Polynomial . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.3 Dierentiation Based on Newtons Backward Interpolation
Polynomial . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.4 Dierentiation Based on Stirlings Interpolation Formula . . .
7.5 Dierentiation Based on Lagranges Interpolation Polynomial
7.6 Two-point and Three-point Formulae . . . . . . . . . . . . . .
7.6.1 Error analysis and optimum step size . . . . . . . . . .
7.7 Richardsons Extrapolation Method . . . . . . . . . . . . . .
7.8 Cubic Spline Method . . . . . . . . . . . . . . . . . . . . . . .
7.9 Determination of Extremum of a Tabulated Function . . . . .
7.10 Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.11 General Quadrature Formula Based on Newtons Forward
Interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.11.1 Trapezoidal Rule . . . . . . . . . . . . . . . . . . . . .
7.11.2 Simpsons 1/3 rule . . . . . . . . . . . . . . . . . . . .
7.11.3 Simpsons 3/8 rule . . . . . . . . . . . . . . . . . . . .
7.11.4 Booles rule . . . . . . . . . . . . . . . . . . . . . . . .
7.11.5 Weddles rule . . . . . . . . . . . . . . . . . . . . . . .
7.12 Integration Based on Lagranges Interpolation . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

365
365
368
372
374
374
375
380
380
381
390
392
394
401

403
. . . . . . 403
. . . . . . 403
. . . . . . 404
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

407
410
413
419
420
424
430
431
432

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

433
433
438
445
446
446
448

## xvi Numerical Analysis

7.13 Newton-Cotes Integration Formulae (Closed type)
7.13.1 Some results on Cotes coecients . . . . . .
7.13.2 Deduction of quadrature formulae . . . . .
7.14 Newton-Cotes Formulae (Open Type) . . . . . . .
7.15 Gaussian Quadrature . . . . . . . . . . . . . . . . .
7.15.1 Gauss-Legendre integration methods . . . .
7.15.2 Lobatto integration methods . . . . . . . .
7.15.3 Radau integration methods . . . . . . . . .
7.15.4 Gauss-Chebyshev integration methods . . .
7.15.5 Gauss-Hermite integration methods . . . .
7.15.6 Gauss-Laguerre integration methods . . . .
7.15.7 Gauss-Jacobi integration methods . . . . .
7.16 Euler-Maclaurins Sum Formula . . . . . . . . . . .
7.17 Rombergs Integration . . . . . . . . . . . . . . . .
7.18 Double Integration . . . . . . . . . . . . . . . . . .
7.18.1 Trapezoidal method . . . . . . . . . . . . .
7.18.2 Simpsons 1/3 method . . . . . . . . . . . .
7.19 Monte Carlo Method . . . . . . . . . . . . . . . . .
7.19.1 Generation of random numbers . . . . . . .
7.20 Worked out Examples . . . . . . . . . . . . . . . .
7.21 Exercise . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

449
450
452
453
455
456
462
464
466
468
469
470
473
480
486
486
490
492
495
497
502

## 8 Ordinary Dierential Equations

8.1 Taylors Series Method . . . . . . . . . . . . . . . . . . . . . . . . .
8.2 Picards Method of Successive Approximations . . . . . . . . . . .
8.3 Eulers Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.3.1 Geometrical interpretation of Eulers method . . . . . . . .
8.4 Modied Eulers Method . . . . . . . . . . . . . . . . . . . . . . . .
8.4.1 Geometrical interpretation of modied Eulers method . . .
8.5 Runge-Kutta Methods . . . . . . . . . . . . . . . . . . . . . . . . .
8.5.1 Second-order Runge-Kutta method . . . . . . . . . . . . . .
8.5.2 Fourth-order Runge-Kutta Method . . . . . . . . . . . . . .
8.5.3 Runge-Kutta method for a pair of equations . . . . . . . . .
8.5.4 Runge-Kutta method for a system of equations . . . . . . .
8.5.5 Runge-Kutta method for second order dierential equation
8.5.6 Runge-Kutta-Fehlberg method . . . . . . . . . . . . . . . .
8.5.7 Runge-Kutta-Butcher method . . . . . . . . . . . . . . . . .
8.6 Predictor-Corrector Methods . . . . . . . . . . . . . . . . . . . . .
8.6.1 Adams-Bashforth-Moulton methods . . . . . . . . . . . . .
8.6.2 Milnes method . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

511
513
515
517
518
520
522
526
526
528
533
537
538
539
540
541
541
547

8.7

## Finite Dierence Method . . . . . . . . . . . . . . .

8.7.1 Second order initial value problem (IVP) . .
8.7.2 Second order boundary value problem (BVP)
8.8 Shooting Method for Boundary Value Problem . . .
8.9 Finite Element Method . . . . . . . . . . . . . . . .
8.10 Discussion About the Methods . . . . . . . . . . . .
8.11 Stability Analysis . . . . . . . . . . . . . . . . . . . .
8.11.1 Model dierential problem . . . . . . . . . . .
8.11.2 Model dierence problem . . . . . . . . . . .
8.11.3 Stability of Eulers method . . . . . . . . . .
8.11.4 Stability of Runge-Kutta methods . . . . . .
8.11.5 Stability of Finite dierence method . . . . .
8.12 Exercise . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

## 9 Partial Dierential Equations

9.1 Finite-Dierence Approximations to Partial Derivatives
9.2 Parabolic Equations . . . . . . . . . . . . . . . . . . . .
9.2.1 An explicit method . . . . . . . . . . . . . . . . .
9.2.2 Crank-Nicolson implicit method . . . . . . . . .
9.3 Hyperbolic Equations . . . . . . . . . . . . . . . . . . .
9.3.1 Implicit dierence methods . . . . . . . . . . . .
9.4 Elliptic Equations . . . . . . . . . . . . . . . . . . . . .
9.4.1 Iterative methods . . . . . . . . . . . . . . . . . .
9.5 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.6 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 Least Squares Approximation
10.1 General Least Squares Method . . . . . . . . .
10.2 Fitting of a Straight Line . . . . . . . . . . . .
10.3 Fitting of a Parabolic Curve . . . . . . . . . . .
10.4 Fitting of a Polynomial of Degree k . . . . . . .
10.5 Fitting of Other Curves . . . . . . . . . . . . .
10.5.1 Geometric curve . . . . . . . . . . . . .
10.5.2 Rectangular hyperbola . . . . . . . . . .
10.5.3 Exponential curve . . . . . . . . . . . .
10.6 Weighted Least Squares Method . . . . . . . .
10.6.1 Fitting of a weighted straight line . . . .
10.7 Least Squares Method for Continuous Data . .
10.8 Approximation Using Orthogonal Polynomials .
10.9 Approximation of Functions . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

Contents

xvii

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

552
553
555
560
563
571
572
572
572
573
575
577
577

.
.
.
.
.
.
.
.
.
.

583
585
586
586
588
597
599
600
605
615
617

.
.
.
.
.
.
.
.
.
.
.
.
.

619
619
620
623
625
626
626
627
628
628
628
630
633
636

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

## xviii Numerical Analysis

10.9.1 Chebyshev polynomials . . . . . . . . . . . . . . . .
10.9.2 Expansion of function using Chebyshev polynomials
10.9.3 Economization of power series . . . . . . . . . . . . .
10.10Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

637
640
645
646

## List of Algorithms and Programs

Sl. No.
3.1
3.2
3.3
3.4
3.5
3.6
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9

Description
Lagranges interpolation for single variable
Newtons forward interpolation
Aitkens interpolation
Interpolation by divided dierence
Interpolation by cubic spline
Lagrange bivariate interpolation
Solution of an equation by bisection method
Solution of an equation by Regula-Falsi method
Solution of an equation by xed point iteration method
Solution of an equation by Newton-Raphson method
Solution of an equation by secant method
Roots of polynomial equation by Birge-Virta method
Roots of polynomial equation by Bairstow method
Seidal iteration method for a pair of non-linear equations
Newton-Rapshon method for a pair of equations
Determinant using partial pivoting
Determinant using complete pivoting
Determination of matrix inverse
Solution of a system of equations by matrix inverse
method
Solution of a system of equations by Gauss elimination
method
Solution of a system of equations by LU decomposition
method
Solution of a tri-diagonal system of equations
Solution of a system of equations by Gauss-Jacobis iteration
Solution of a system of equations by Gauss-Seidals iteration
xix

Algo.
85
93
115
129
149
162
197
200
207
224
228
243
250
264
268
282
284
290
294

Prog.
85
94
115
130
151
163
197
201
208
225
229
244
251
264
269
283
285
291
294

300

301

310

311

323
341

323
342

350

351

xx Numerical Analysis
Sl. No.
5.20
6.1
6.2
6.3
6.4
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
8.1
8.2
8.3
8.4
8.5
8.6
8.7
9.1
9.2
10.1
10.2

Description
Solution of a system of equations by Gauss-Seidal SOR
method
Characteristic polynomial of a matrix by LeverrierFaddeev method
Largest eigenvalue by power method
Eigenvalue of a real symmetric matrix by Jacobis method
Eigenvalue of a real symmetric matrix by Householder
method
First derivative based on Lagranges interpolation
First derivative using Richardson extrapolation
Integration by trapezoidal rule
Integration by Simpsons 1/3 rule
Rombergs integration
Double integration using trapezoidal rule
Integration by Monte Carlo method
Solution of a rst order dierential equation by Eulers
method
Solution of a rst order dierential equation by modied
Eulers method
Solution of a rst order dierential equation by fourth
order Runge-Kutta method
Solution of a pair of rst order dierential equation by
Runge-Kutta method
Solution of a rst order dierential equation by AdamsBashforth-Moulton method
Solution of a rst order dierential equation Milnes
predictor-corrector method
Solution of a second order BVP using nite dierence
method
Solution of heat equation using Crank-Nicolson implicit
method
Solution of Poissons equation using Gauss-Seidal S.O.R.
method
Fitting of straight line by least square method
Approximation of a function by Chebyshev polynomial

Algo.

Prog.
357

370

370

378
386
398

379
387
399

415
428
437
444
461
483
488
494
519

416
429
438
444
461
484
489
496
519

524

524

531

532

535

536

544

545

550

550

557

557

593

594

612

612

622
642

623
643

Chapter 1

Errors in Numerical
Computations
The solutions of mathematical problems are of two types: analytical and numerical.
The analytical solutions can be expressed in closed form and these solutions are error
free. On the other hand, numerical method is a division of mathematics which solves
problems using computational machine (computer, calculator, etc.). But, for some
classes of problems it is very dicult to obtain an analytical solution. For example, the
Indian populations are known at the years 1951, 1961, 1971, 1981, 1991, 2001. There
is no analytical method available to determine the population in the year, say, 2000.
But, using numerical method one can determine the population in the said year. Again,
sometimes we observed that the solutions of non-linear dierential equations cannot be
determined by analytical methods, but, such problems can easily be solved by numerical
methods. Numerical computations are almost invariably contaminated by errors, and
it is important to understand the source, propagation, magnitude, and rate of growth
of these errors.
In this age of computer, many complicated and large problems are solved in signicantly less time. But, without using numerical methods we cannot solve any mathematical problem using computer, as analytical methods are not suitable to solve a problem
by computer. Thus, the numerical methods are highly appreciated and extensively used
by Mathematicians, Computer Scientists, Statisticians, Engineers and others.

1.1

Sources of Errors

## The solution of a problem obtained by numerical method contains some errors. To

minimize the errors, it is most essential to identify the causes or sources of the errors
1

2 Numerical Analysis
and their growth and propagation in numerical computation. Three types of errors, viz.,
inherent errors, round-o errors and truncation errors, occur in nding the solution of
a problem using numerical method. These three type of errors are discussed below.
(i) Inherent errors: This type of errors is present in the statement of the problem itself,
before determining its solution. Inherent errors occur due to the simplied assumptions
made in the process of mathematical modelling of a problem. It can also arise when the
data is obtained from certain physical measurements of the parameters of the proposed
problem.
(ii) Round-o errors: Generally, the numerical methods are carried out using calculator or computer. In numerical computation, all the numbers are represented by
decimal fraction. Some numbers such as 1/3, 2/3, 1/7 etc. can not be represented by
decimal fraction in nite numbers of digits. Thus, to get the result, the numbers should
be rounded-o into some nite number of digits.
Again, most of the numerical computations are carried out using calculator and computer. These machines can store the numbers up to some nite number of digits. So in
arithmetic computation, some errors will occur due to the nite representation of the
numbers; these errors are called round-o error. Thus, round-o errors occur due to the
nite representation of numbers during arithmetic computation. These errors depend
on the word length of the computational machine.
(iii) Truncation errors: These errors occur due to the nite representation of an
inherently innite process. For example, the use of a nite number of terms in the
innite series to compute the value of cos x, sin x, ex , etc.
The Taylors series expansion of sin x is
sin x = x

x3 x5 x7
+

+ .
3!
5!
7!

This is an innite series expansion. If only rst ve terms are taken to compute the
value of sin x for a given x, then we obtain an approximate result. Here, the error occurs
due to the truncation of the series. Suppose, we retain the rst n terms, the truncation
error (Etrunc ) is given by
Etrunc

x2n+1
.
(2n + 1)!

It may be noted that the truncation error is independent of the computational machine.

1.2

## Exact and Approximate Numbers

To solve a problem, two types of numbers are used. They are exact and approximate.
Exact number gives a true value of a result and approximate number gives a value which
is closed to the true value.

## Errors in Numerical Computations

For example, in the statements a triangle has three sides, there are 2000 people in a
locality, a book has 450 pages the numbers 3, 2000 and 450 are exact numbers. But,
in the assertions the height of a pupil is 178 cm, the radius of the Earth is 6400 km,
the mass of a match box is ten gram, the numbers 178, 6400 and 10 are approximate
numbers.
This is due to the imperfection of measuring instruments we use. There are no
absolutely exact measuring instruments; each of them has its own accuracy. Thus, the
height of a pupil is 178 cm is not absolute measurement. In the second example, the
radius of the Earth is very concept; actually, the Earth is not a sphere at all, and we
can use its radius only in approximate terms. In the last example, the approximation
of the number is also dened by the fact that dierent boxes may have dierent masses
and the number 10 denes the mass of a particular box.
One important observation is that, same number may be exact as well as approximate.
For example, the number 3 is exact when it represents the number of sides of a triangle
and approximate if we use it to represent the number when calculating the area of a
circle using the formula r2 .

## Independently, the numbers 1, 2, 3, 12 , 53 , 2, , e, etc. written in this manner are exact.

An approximate value of is 3.1416, a better approximation of it is 3.14159265. But
one cannot write the exact value of .
The accuracy of calculations is dened by the number of digits in the result which
enjoy condence. The signicant digits or signicant gures of a number are all its
digits, except for zeros which appear to the left of the rst non-zero digit. Zeros at the
end of a number are always signicant digit. For example, the numbers 0.001205 and
356.800 have 4 and 6 signicant digits respectively.
In practical calculations, some numbers occur containing large number of digits, and
it will be necessary to cut them to a usable number of gures. This process is called
rounding-o of numbers. That is, in rounding process the number is replaced by
another number consisting of a smaller number of digits. In that case, one or several
digits keep with the number, taken from left to right, and discard all others.
The following rules of rounding-o are commonly used:
(i) If the discarded digits constitute a number which is larger than half the unit in the
last decimal place that remains, then the last digit that is left is increased by one.
If the discarded digits constitute a number which is smaller than half the unit in
the last decimal place that remains, then the digits that remain do not change.
(ii) If the discarded digits constitute a number which is equal to half the unit in the
last decimal place that remains, then the last digit that is half is increased by one,
if it is odd, and is unchanged if it is even.

4 Numerical Analysis
This rule is often called a rule of an even digit. If a number is rounded using the
above rule then the number is called correct up to some (say n) signicant gures.
The following numbers are rounded-o correctly to ve signicant gures:
Exact number
25.367835
28.353215
3.785353
5.835453
6.73545
4.83275
0.005834578
3856754
2.37
8.99997
9.99998

Round-o number
25.368
28.353
3.7854
5.8355
6.7354
4.8328
0.0058346
38568102
2.3700
9.0000
10.000

From above examples, it is easy to observe that, while rounding a number, an error
is generated and this error is sometimes called round-o error.

1.3

## Absolute, Relative and Percentage Errors

Let xT be the exact value of a number and xA be its approximate value. If xA < xT ,
then we say that the number xA is an approximate value of the number xT by defect
and if xA > xT , then it is an approximate value of xT by excess.
The dierence between the exact value xT and its approximate value xA is an error.
As a rule, it is not possible to determine the value of the error xT xA and even its
sign, since the exact number xT is unknown.
The errors are represented in three ways, viz., absolute error, relative error and
percentage error.
Absolute error:
The absolute error of the approximate number xA is a quantity (x) which satises the
inequality
x |xT xA |.
The absolute error is the upper bound of the deviation of the exact number xT from
its approximation, i.e.,
xA x xT xA + x.
The above result can be written in the form
xT = xA x.

(1.1)

## Errors in Numerical Computations

In other words, the absolute error of the number x is the dierence between true
value and approximate value, i.e.,
x = |xT xA |.
It may be noted from the rounding process that, if a number be rounded to m decimal
places then
1
absolute error 10m .
(1.2)
2
The absolute error measures only the quantitative aspect of the error but not the
qualitative one, i.e., does not show whether the measurement and calculation were
accurate. For example, the length and the width of a table are measured with a scale
(whose division is 1 cm) and the following results are obtained: the width w = 5 0.5
cm and the length l = 100 0.5 cm. In both cases the absolute error is same and it is
0.5 cm. It is obvious that the second measurement was more accurate than the rst.
To estimate the quality of calculations or measurements, the concept of a relative error
is introduced.
Relative error:
The relative error (x) of the number xA is
x =

x
x
or
, |xT | = 0 and |xA | = 0.
|xA |
|xT |

## This expression can be written as

xT = xA (1 x) or xA = xT (1 x).
Note that relative error is the absolute error when measuring 1 unit.
For the measurements of the length and the width of the table (discussed earlier) the
relative errors are
0.5
0.5
= 0.1 and l =
= 0.005.
w =
5
100
In these cases, one can conclude that the measurement of the length of the table has
been relatively more accurate than that of its width. So one conclusion can be drawn:
the relative error measures the quantity and quality of the calculation and measurement.
Thus, the relative error is a better measurement of error than absolute error.
Percentage error:
The percentage error of an approximate number xA is x 100%.
It is a particular type of relative error. This error is sometimes called relative
percentage error. The percentage error gives the total error while measuring 100 unit
instead of 1 unit. This error also calculates the quantity and quality of measurement.
When relative error is very small then the percentage error is calculated.

6 Numerical Analysis
Note 1.3.1 The absolute error of a number correct to n signicant gures cannot be
greater than half a unit in the nth place.

Note 1.3.2 The relative error and percentage error are independent of the unit of
measurement, while absolute error depends on the measuring unit.

## Dierence between relative error and absolute error:

Absolute error measures only quantity of error and it is the total amount of error
incurred by approximate value. While the relative error measures both the quantity
and quality of the measurement. It is the total error while measuring one unit. The
absolute error depends on the measuring unit, but, relative error does not depend on
measuring unit.
1
Example 1.3.1 Find the absolute, relative and percentage error in xA when xT =
3
and xA = 0.333.
Solution. The absolute error
x = |xT xA | =
=

1 0.999
1
0.333 =
3
3

0.001
= 0.00033.
3

## The relative error

x =

x
0.00033
= 0.00099  0.001.
=
xT
1/3

## The percentage error is x 100% = 0.00099 100% = 0.099%  0.1%.

Example 1.3.2 An exact number xT is in the interval [28.03, 28.08]. Assuming an
approximate value, nd the absolute and the percentage errors.
Solution. The middle of the given interval is taken as its approximate value, i.e.,
xA = 28.055. The absolute error is half of its length, i.e., x = 0.025. The relative
x
= 0.000891 .
error x =
xA
It is conventional to round-o the error to one or two non-zero digits. Therefore,
x = 0.0009 and the percentage error is 0.09%.

## Errors in Numerical Computations

Example 1.3.3 Determine the absolute error and the exact number corresponding
to the approximate number xA = 5.373 if percentage error is 0.01%.
Solution. Here the relative error x = 0.01% = 0.0001.
The absolute error x = |xA x| = 5.373 0.0001 = 0.0005373  0.00054.
The exact value = 5.373 0.00054.
Example 1.3.4 Find out in which of the following cases, the quality of calculations

15
 0.8824 and yT = 51  7.141.
is better: xT =
17
Solution. To nd the absolute error, we take thenumbers xA and yA with a larger
number of decimal digits as xA  0.882353, yA = 51  7.141428.
Therefore, the absolute error in xT is |0.882353 0.8824|  0.000047,
and in yT is |7.141428 7.141|  0.00043.
The relative error in xA is 0.00047/0.8824  0.00053 = 0.05%
and relative error in yA is 0.00043/7.141 = 0.0000602 = 0.006%.
In the second case the quality of calculation is better than the rst case as relative
error in xT > relative error in yT .

1.4

## Valid Signicant Digits

A real number can be represented by many dierent ways. For example, the number
840000 can be represented as two factors: 840 103 or 84.0 104 or 0.840 106 . (Note
that in these representations the last three signicant zeros are lost). The later form of
the notation is known as normalize form and it is commonly used. In this case, we
say that 840 is the mantissa of the number and 6 is its order.
Every positive decimal number, exact as well as approximate, can be expressed as
a = d1 10m + d2 10m1 + + dn 10mn+1 + ,
where di are the digits constituting the number (i = 1, 2, . . .) with d1 = 0 and 10mi+1
is the value of the ith decimal position (counting from left).
The digit dn of the approximate number a is valid signicant digit (or simply a
valid digit) if it satises the following inequality.
a 0.5 10mn+1 ,

(1.3)

i.e., absolute error does not exceed half the unit of the decimal digit in which dn appears.
If inequality (1.3) is not satised, then the digit dn is said to be doubtful. It is
obvious that if the digit dn is valid, then all the preceding digits, to the left of it, are
also valid.

8 Numerical Analysis
Theorem 1.1 If a number is correct up to n signicant gures and the rst signicant
digit of the number is k, then the relative error is less than
1
.
k 10n1
Proof. Let xA be the approximate value of the exact number xT . Also, let xA is correct
up to n signicant gures and m decimal places. Then there are three possibilities may
occur:
(i) m < n
(ii) m = n and
(iii) m > n.
We have by (1.2), the absolute error x 0.5 10m .
Case I. When m < n.
In this case, the total number of digits in integral part is n m. If k be the rst
signicant digit in xT , then
x 0.5 10m and |xT | k 10nm1 0.5 10m .
Therefore, the relative error
x =

0.5 10m
x

|xT |
k 10nm1 0.5 10m
1
.
=
2k 10n1 1

## Since, n is a positive integer and k is an integer lies between 1 and 9,

2k 10n1 1 > k 10n1
for all k and n except k = n = 1.
Hence,
x <

1
.
k 10n1

## Case II. When m = n.

In this case, the rst signicant digit is the rst digit after decimal point, i.e., the
integral part is zero.
As before,
0.5 10m
x =
k 10nm1 0.5 10m
1
1
<
=
.
2k 10n1 1
k 10n1

## Case III. When m > n.

In this case, the rst signicant digit k is at the (n m + 1) = (m n 1)th position.
Also, the integer part is zero. Then x 0.5 10m and |xT | k 10(mn+1)
0.5 10m .
Therefore,
0.5 10m
x =
k 10(mn+1) 0.5 10m
1
1
<
=
.
n1
2k 10
1
k 10n1
Hence the theorem.

1.5
1.5.1

## Propagation of Errors in Arithmetic Operations

The errors in sum and dierence

## Consider the exact numbers X1 , X2 , . . . , Xn and their approximations be respectively

x1 , x2 , . . . , xn . Let x1 , x2 , . . . , xn be the errors in x1 , x2 , . . . , xn , i.e., Xi = xi
xi , i = 1, 2, . . . , n. Also, let X = X1 + X2 + + Xn and x = x1 + x2 + + xn .
Therefore, the total absolute error is
|X x| = |(X1 x1 ) + (X2 x2 ) + + (Xn xn )|
|X1 x1 | + |X2 x2 | + + |Xn xn |.
Thus the absolute error in the sum is
|x| = |x1 | + |x2 | + + |xn |.

(1.4)

Thus the absolute error in sum of approximate numbers is equal to the sum of the
absolute errors of the numbers.
From (1.4), it follows that the absolute error of the algebraic sum must not be smaller
than the absolute error of the least exact term.
The following points should be kept in mind when adding numbers of dierent absolute accuracy.
(i) identify a number (or numbers) of the least accuracy (i.e., a number which has
the maximum absolute error),
(ii) round-o more exact numbers so as to retain in them one digit more than in the
identied number (i.e., retain one reserve digit),
(iii) perform addition taking into account all the retained digits,
(iv) round-o the result by discarding one digit.

10 Numerical Analysis
Subtraction
Consider x1 and x2 be two approximate values of the corresponding exact numbers X1
and X2 . Let X = X1 X2 and x = x1 x2 .
Then X1 = x1 x1 and X2 = x2 x2 , where x1 and x2 are the errors in x1
and x2 respectively.
Therefore, |X x| = |(X1 x1 ) (X2 x2 )| |X1 x1 | + |X2 x2 |. Hence,
|x| = |x1 | + |x2 |.

(1.5)

Thus the absolute error in dierence of two numbers is equal to the sum of individual
absolute errors.
1.5.2

## The error in product

Let us consider two exact numbers X1 and X2 and their approximate values x1 and x2 .
Also, let x1 and x2 be the errors in x1 and x2 , i.e., X1 = x1 x1 and X2 = x2 x2 .
Now, X1 X2 = x1 x2 x1 x2 x2 x1 x1 x2 .
Then |X1 X2 x1 x2 | |x1 x2 | + |x2 x1 | + |x1 x2 |. The last term of right hand
side is small, so we discard it and dividing both sides by |x| = |x1 x2 |.
Thus the relative error in the product is

 
 

 X1 X2 x1 x2   x2   x1 

=
 

(1.6)

  x2  +  x1 .
x1 x2
Thus the relative errors in product of two numbers is equal to the sum of individual
relative errors.
The result (1.6) can be easily extended to the product of several numbers so that, if
X = X1 X2 Xn and x = x1 x2 xn , then

 



 
 X x   x1   x2 



+
 + +  xn .
=
(1.7)
 x   x1   x2 
 xn 
That is, the total relative error in product of n numbers is equal to the sum of
individual relative errors.
A particular case
Let the approximate numbers x1 , x2 , . . . , xn be all positive and x = x1 x2 xn .
Then log x = log x1 + log x2 + + log xn .
x
x1 x2
xn
=
+
+ +
.
x
x1
x2
xn
 x   x   x 
 x 

 


1
2
n
That is, 
=
+
 + + 
.
x
x1
x2
xn
Usually, the following steps are followed when multiplying two numbers:
Therefore,

11

## (i) identify a number with the least number of valid digits,

(ii) round o the remaining factors so that they would contain one signicant digit
more than the valid signicant digits there are in the isolated number,
(iii) retain as many signicant digits in the product as there are valid signicant digits
in the least exact factor (the identied number).
Example 1.5.1 Show that when an approximate number x is multiplied by an
exact factor k, the relative error of the product is equal to the relative error of the
approximate number x and the absolute error is |k| times the absolute error of the
absolute number.
Solution. Let x = kx1 , where k is an exact factor other than zero. Then
 


 
 x   k x1   x1 
=
 = x1 .
=
x = 
x   k x1   x1 
But, the absolute error |x| = |k x1 | = |k| |x1 | = |k| times the absolute error in
x1 .
1.5.3

## The error in quotient

Let us consider two exact numbers X1 and X2 and their approximate values x1 and x2 .
x1
X1
and x =
.
Also, let X =
X2
x2
Then X1 = x1 + x1 , X2 = x2 + x2 , where x1 and x2 are the errors.
Let x1 = 0 and x2 = 0.
Now,
x2 x1 x1 x2
x1 + x1 x1
.

=
X x=
x2 + x2 x2
x2 (x2 + x2 )
Dividing both sides by x and taking absolute values:



 
 
 X x   x2 x1 x1 x2  
  x1 x2 
x2



=
=
 x   x1 (x2 + x2 )   x2 + x2   x1 x2 .
The error x2 is small as compared to x2 , then approximately
x2
 1. Therefore, above relation becomes
x2 + x2
 
 

 
 

 x   X x   x1 x2   x1   x2 


,








x = 

+
=
=
x   x   x1
x2   x1   x2 

(1.8)

i.e., x = x1 + x2 . Hence, the total relative error in quotient is equal to the sum of
their individual relative errors.

12 Numerical Analysis
The relation (1.8) can also be written as
 
 


 
 x   x1 x2   x1   x2 






=
 x   x1 x2   x1   x2 .

(1.9)

From this relation one can conclude that the relative error in quotient is greater than
or equal to the dierence of their individual relative errors.
A particular case
For positive approximate numbers x1 and x2 , the equation (1.8) can easily be deduced.
Let x = x1 /x2 . Then log x = log x1 log x2 . Therefore,

 

 
 x   x1   x2 
x1 x2
x



.


=

i.e., 
+

x
x1
x2
x   x1   x2 
While dividing two numbers the following points should be followed.
(i) identify the least exact number, i.e., the number with the least number of valid
digits,
(ii) round-o the other number, leaving in it on signicant digit more than there are
digits in the identied number,
(iii) retain as many signicant digits in the quotient as there were in the least exact
number.
Example 1.5.2 Find the sum of the approximate numbers 0.543, 0.1834, 17.45,
0.000234, 205.2, 8.35, 185.3, 0.0863, 0.684, 0.0881 in each of which all the written
digits are valid. Find the absolute error in sum.
Solution. The least exact numbers (those possessing the maximum absolute error)
are 205.2 and 185.3. The error of each of them is 0.05. Now, rounding o the other
numbers, leaving one digit more and adding all of them.
0.54+0.18+17.45+0.00+205.2+8.35+185.3+0.09+0.68+0.09=417.88.
Discarding one digit by round-o the sum and we obtained 417.9.
The absolute error in the sum consists of two terms:
(i) the initial error, i.e., the sum of the errors of the least exact numbers and the
rounding errors of the other numbers: 0.05 2 + 0.0005 8 = 0.104  0.10.
(ii) the error in rounding-o the sum is 417.9 417.88 = 0.02.
Thus the absolute error of the sum is 0.10 + 0.02 = 0.12.
So, the sum can be written as 417.9 0.12.

## Errors in Numerical Computations

13

Example 1.5.3 Find the dierence of the approximate numbers 27.5 and 35.8 having absolute errors 0.02 and 0.03 respectively. Evaluate the absolute and the relative
errors of the result.
Solution. Let x1 = 27.5 and x2 = 35.8. Then x = x1 x2 = 8.3. The total
absolute error x = 0.02 + 0.03 = 0.05.
Thus the dierence x1 x2 is 8.3 with absolute error 0.05.
The relative error is 0.05/| 8.3|  0.006 = 0.6%.
Example 1.5.4 Find the product of the approximate numbers x1 = 8.6 and x2 =
34.359 all of whose digits are valid. Also nd the relative and the absolute errors.
Solution. In the rst number, there are two valid signicant digits and in the second
there are ve digits. Therefore, round-o the second number to three signicant
digits. After rounding-o the numbers x1 and x2 become x1 = 8.6 and x2 = 34.4.
Hence the product is
x = x1 x2 = 8.6 34.4 = 295.84  3.0 102 .
In the result, there are two signicant digits, because the least number of valid signicant digits of the given numbers is 2.
The relative error in product is
 

 

 x   x1   x2  0.05 0.0005


=



+
= 0.00583  0.58%.
+
=
x = 
x   x1   x2 
8.6
34.359
The absolute error is (3.0 102 ) 0.00583 = 1.749  1.7.
Example 1.5.5 Calculate the quotient x/y of the approximate numbers x = 6.845
and y = 2.53 if all the digits of the numbers are valid. Find the relative and the
absolute errors.
Solution. Here the dividend x = 6.845 has four valid signicant digits and the
divisor has three, so we perform division without rounding-o. Thus
6.845
x
=
= 2.71.
y
2.53
Three signicant digits are retained in the result, since, the least exact number (the
divisor y) contains three valid signicant digits.
The absolute error in x and y are respectively
x = 0.0005 and y = 0.005.

14 Numerical Analysis
Therefore the relative error in quotient is

 

 x   y  0.0005 0.005

+

 x   y  = 6.845 + 2.53 = 0.000073 + 0.00198
 0.002 = 0.2%.
The absolute error is
 
x
  0.002 = 2.71 0.002 = 0.00542 = 0.005.
y

1.5.4

## The errors in power and in root

Let us consider an approximate number x1 which has a relative error x1 . Now, the
problem is to nd the relative error of x = xm
1 .
Then
x1 x1 x1 .
x = xm
1 =


m times

## By (1.7), the relative error x in the product is

x = x1 + x1 + + x1 = m x1 .




(1.10)

m times

Thus, when the approximate number x is raised to the power m, its relative error
increases m times.

## Similarly, one can calculate the relative error of the number x = m x1 .

Here x1 > 0. Therefore,
1
log x1 .
log x =
m
That is,





 x 
x
1 x1
1  x1 


or 
=
.
=
x
m x1
x  m  x1 

x =

1
x1 ,
m

## Errors in Numerical Computations

15

X3 Y
Example 1.5.6 Calculate A =
where X = 8.36, Y = 80.46, Z = 25.8. The
Z2
absolute errors in X, Y, Z are respectively 0.01, 0.02 and 0.03. Find the error of the
result.
Solution. Here the absolute error x = 0.01, y = 0.02 and z = 0.03. To calculate
intermediate result, retain one reserve digit. The approximate intermediate values

## are x3 = 584.3, y = 8.9699, z 2 = 665.6, where x, y, z are approximate values of

X, Y, Z respectively.
Thus the approximate value of the expression is
a=

584.3 8.9699
= 7.87.
665.6

Three signicant digits are taken in the result, since, the least number of signicant
digits in the numbers is 3.
Now, the relative error a in a is given by
0.01 1
0.02
0.03
1
y + 2 z = 3
+
+2
2
8.36 2 80.46
25.8
 0.0036 + 0.00012 + 0.0023  0.006 = 0.6%.

a = 3 x +

## The absolute error a in a is 7.87 0.006 = 0.047.

Hence, A = 7.87 0.047 and the relative error is 0.006.
1.5.5

## Let y = f (x1 , x2 , . . . , xn ) be a dierentiable function containing n variables x1 , x2 , . . . , xn

and let xi be the error in xi , for all i = 1, 2, . . . , n.
Then the error y in y is given by
y + y = f (x1 + x1 , x2 + x2 , . . . , xn + xn )
n

f
xi +
= f (x1 , x2 , . . . , xn ) +
xi
i=1

n

f
xi
=y+
xi
i=1

## (neglecting second and higher powers terms of xi )

n

f
i.e., y =
xi
xi
i=1

16 Numerical Analysis
This formula gives the total error for computing a function containing several variables.
The relative error is given by
y  f xi
=
.
y
xi y
n

i=1

1.6

Signicant Error

Signicant error occurs due to the loss of signicant digits during arithmetic computation. This error occurs mainly due to the nite representation of the numbers in
computational machine (computer or calculator). The loss of signicant digits occurs
due to the following two reasons:
(i) when two nearly equal numbers are subtracted and
(ii) when division is made by a very small divisor compared to the dividend.
Signicant error is more serious than round-o error, which are illustrated in the
following examples:

Example 1.6.1 Find the dierence X = 5.36 5.35 and evaluate the relative
error of the result.

## Solution. Let X1 = 5.36  2.315 = x1 and X2 = 5.35  2.313 = x2 .

The absolute errors x1 = 0.0005 and x2 = 0.0005. Then the approximate dierence is x = 2.315 2.313 = 0.002.
The total absolute error in the subtraction is x = 0.0005 + 0.0005 = 0.001.
0.001
= 0.5 = 50%.
The relative error x =
0.002
However, by changing the scheme of calculation we get a more accurate result.

## ( 5.36 5.35)( 5.36 + 5.35)

X = 5.36 5.35 =
5.36 + 5.35
0.01
5.36 5.35

=
 0.002 = x (say).
=
5.36 + 5.35
5.36 + 5.35
In this case the relative error is
x =

0.001
x1 + x2
= 0.0002 = 0.02%.
=
x1 + x2
2.315 + 2.313

Thus, when calculating x1 and x2 with the same four digits we get a better result in
the sense of a relative error.

## Errors in Numerical Computations

17

Example 1.6.2 Calculate the values of the function y = 1 cos x at x = 82o and
at x = 1o . Also, calculate the absolute and the relative errors of the results.
Solution. y at x = 82o
The value of cos 82o  0.1392 = a1 (say) (correct up to four digits) and a1 =
0.00005. Then y1 = 1 0.1392 = 0.8608 and y1 = 0.00005 (from an exact number equal to unity we subtract an approximate number with an absolute error not
exceeding 0.00005).
Consequently, the relative error is
y1 =

0.00005
= 0.000058 = 0.006%.
0.8608

y at x = 1o
We have cos 1o  0.9998 = a2 (say). a2 = 0.00005.
y2 = 1 0.9998 = 0.0002. y2 = 0.00005.
Hence
0.00005
= 0.25 = 25%.
y2 =
0.0002
From this example it is observed that for small values of x, a direct calculation of
y = 1 cos x gives a relative error of the order 25%. But at x = 82o the relative error
is only 0.006%.
Now, change the calculation procedure and use the formula y = 1 cos x = 2 sin2 x2
to calculate the value of y for small values of x.
Let a = sin 0o 30  0.0087. Then a = 0.00005 and
a =

0.00005
= 0.0058 = 0.58%.
0.0087

## Thus y2 = 2 0.00872 = 0.000151 and relative error

y2 = 0.0058 + 0.0058 = 0.012 = 1.2% (using the formula a = x + y if a = x.y).
The absolute error is
y2 = y2 y2 = 0.000151 0.012 = 0.000002.
Thus a simple transformation, of the computing formula, gives a more accurate result
for the same data.
Example 1.6.3 Find the roots of the equation x2 1000x + 0.25 = 0.
Solution. For simplicity, it is assumed that all the calculations are performed using
four signicant digits. The roots of this equation are

1000 106 1
.
2

18 Numerical Analysis
Now,
106 1 = 0.1000 107 0.0000 107 = 0.1000 107 .
Thus 106 1 = 0.1000 104 .
0.1000 104 0.1000 104
Therefore the roots are
2
which are respectively 0.1000 104 and 0.0000 104 . One of the roots becomes zero
due to the nite representation of the numbers. But, the transformed formula gives
the smaller root more accurately.
The smaller root of the equation may be calculated using the transformed formula

## (1000 106 1)(1000 + 106 1)

1000 106 1

=
2
2(1000 + 106 1)
1

=
= 0.00025.
2(1000 + 106 1)
Thus the roots of the given equation are 0.1000 104 and 0.00025.
Such a situation may be recognized by checking |4ac|  b2 .
It is not always possible to transform the computing formula. Therefore, when nearly
equal numbers are subtracted, they must be taken with a sucient number of reserve
valid digits. If it is known that the rst m signicant digits may be lost during computation and if we need a result with n valid signicant digits then the initial data should
be taken with m + n valid signicant digits.

1.7

## Today, generally, numerical computations are carried out by calculator or computer.

Due to the limitations of calculator, computers are widely used in numerical computation. In this book, computer is taken as the computational machine. Now, the
representation and computation of numbers in computer are discussed below.
In computer, the numbers are stored mainly in two forms: (i) integer or xed point
form, and (ii) real or oating point form. Before storing to the computer, all the numbers
are converted into binary numbers (consisting two bits 0 and 1) and then these converted
numbers are stored into computer memory. Generally, two bytes (two bytes equal to 16
bits, one bit can store either 0 or 1) memory space is required to store an integer and
four bytes space is required to store a oating point number. So, there is a limitation
to store the numbers into computers.
Storing of integers is straight forward while representation of oating point numbers
is dierent from our conventional technique. The main aim of this new technique is to
preserve the maximum number of signicant digits in a real number and also increase
the range of values of the real numbers. This representation is called the normalized

## Errors in Numerical Computations

19

oating point mode. In this mode of representation, the whole number is converted to
a proper fraction in such a way that the rst digit after decimal point should be non-zero
and is adjusted by multiplying a number which is some powers of 10. For example, the
number 375.3 104 is represented in this mode as .3753 107 = .3753E7 (E7 is used
to represent 107 ). From this example, it is observed that in normalized oating point
representation, a number is a combination of two parts mantissa and exponent.
In the above example, .3753 is the mantissa and 7 is the exponent. It may be noted
that the mantissa is always greater than or equal to .1 and exponent is an integer.
For simplicity, it is assume that the computer (hypothetical) uses four digits to store
mantissa and two digits for exponent. The mantissa and the exponent have their own
signs.
The number .0003783 would be stored as .3783E3. The leading zeros in this number
serve only to indicate the decimal point. Thus, in this notation the range of numbers
(magnitudes) is .9999 1099 to .1000 1099 .

1.8

## Arithmetic of Normalized Floating Point Numbers

In this section, the arithmetic operations on normalized oating point numbers are
discussed.
1.8.1

If two numbers have same exponent, then the mantissas are added directly and the
If the exponents are dierent then lower exponent is shifted to higher exponent by
Example 1.8.1 Add the following normalized oating point numbers.
(i) .3456E3 and .4325E3 (same exponent)
(ii) .8536E5 and .7381E5
(iii) .3758E5 and .7811E7 (dierent exponent)
(iv) .2538E2 and .3514E7
(v) .7356E99 and .3718E99 (overow condition)
Solution. (i) In this case, the exponents are equal, so the mantissa are added
directly. Thus the sum is .7781E3.
(ii) In this case, the exponent are equal and the sum is 1.5917E5. Here the mantissa
has 5 signicant digits, but our computer (hypothetical) can store only four signicant
gures. So, the number is shifted right one place before it is stored. The exponent is
increased by 1 and the last digit is truncated. The nal result is .1591E6.

20 Numerical Analysis

(iii) Here, the numbers are .3758E5 and .7811E7. The exponent of the rst number
is less than that of the second number. The dierence of the exponents is 7 5 = 2.
So the mantissa of the smaller number (here rst number) is shifted right by 2 places
(the dierence of the exponents) and the last 2 digits of the mantissa are discarded
as our hypothetical computer can store only 4 digits. Then the rst number becomes
.0037E7. Then the result is .0037E7 + .7811E7 = .7848E7.
(iv) Here also the exponents are dierent and the dierence is 7 2 = 5. The mantissa of rst number (smaller exponent) is shifted 5 places and the number becomes
.0000E7. The nal result is .0000E7 + .3514E7 = .3514E7.
(v) Here the numbers are .7356E99 and .3718E99 and they have equal exponent. So
the sum of them is 1.1074E99. In this case mantissa has ve signicant digits. Thus
the mantissa is shifted right and the exponent is increased by 1. Then the exponent
becomes 100. As the exponent cannot store more than two digits, in our hypothetical
computer, the number is larger than the largest number that can be stored in our
computer. This situation is called an overow condition and the machine will give
an error message.

1.8.2

Subtraction

The subtraction is same as addition. In subtraction one positive number and one negative number are added. The following example shows the details about subtraction.
Example 1.8.2 Subtract the normalized oating point numbers indicated below:
(i) .3628E6 from .8321E6
(ii) .3885E5 from .3892E5
(iii) .3253E7 from .4123E6
(iv) .5321E99 from .5382E99.
Solution. (i) Here the exponents are equal, and the result is
.8321E6 .3628E6 = .4693E6.
(ii) Here the result is .3892E5 .3885E5 = .0007E5. The most signicant digit in the
mantissa is 0, so the mantissa is shifted left till the most signicant digit becomes
non-zero and in each left shift of the mantissa the exponent is reduced by 1. Hence
the nal result is .7000E2.
(iii) The numbers are .4123E6 and .3253E7. The exponents are not equal, so the
number with smaller exponent is shifted right and the exponent increased by 1 for
every right shift. Then the second number becomes .0325E6. Thus the result is
.4123E6 .0325E6 = .3798E6.

## Errors in Numerical Computations

21

(iv) The result is .5382E99 .5321E99 = .0061E99. For normalization, the mantissa is shifted left twice and in this process the exponent is reduced by 1. In rst
shift, the exponent becomes 100, but our hypothetical computer can store only two
digits as exponent. So 100 cannot be accommodated in the exponent part of the
number. In this case, the result is smaller than the smallest number which could be
stored in our computer. This condition is called an underow condition and the
computer will give an error message.
1.8.3

Multiplication

Two numbers in normalized oating point mode are multiplied by multiplying the mantissa and adding the exponents. After multiplication, the mantissa is converted into
normalized oating point form and the exponent is converted appropriately. The following example shows the steps of multiplication.
Example 1.8.3 Multiply the following numbers indicated below:
(i) .5321E5 by .4387E10
(ii) .1234E10 by .8374E10
(iii) .1139E50 by .8502E51
(iv) .3721E52 by .3205E-53.
Solution. (i) Here, .5321E5 .4387E10 = .23343227
  E15.

The mantissa has 8 signicant gures, so the last four digits are discarded. The nal
result is .2334E15.
(ii) Here, .1234E10 .8374E10 = .10333516
  E0 = .1033E0.

## (iii) .1139E50 .8502E51 = .09683778E101.

Here, the mantissa has one 0 as most signicant digit, so the mantissa is shifted
left one digit and the exponent is adjusted. The product is .9683E100. But our
hypothetical computer cannot store 3 digits as exponent. Hence, in this case, the
overow condition occurs.
(iv) .3721E52 .3205E53 = .11925805E105 = .1192E105.
In this case, the product is very small (as the exponent is 105). Hence the underow
condition occurs.
1.8.4

Division

In the division, the mantissa of the numerator is divided by that of the denominator.
The exponent is obtained by subtracting exponent of denominator from the exponent

22 Numerical Analysis
of numerator. The quotient mantissa is converted to normalized form and the exponent
Example 1.8.4 Perform the following divisions
(i) .9938E5 .3281E2
(ii) .9999E2 .1230E99
(iii) .3568E10 .3456E97.
Solution. (i) .9938E5 .3281E2 = .3028E4.
(ii) .9999E2 .1230E99 = .8129E102.
The result overows.
(iii) .3568E10 .3456E97 = .1032E106.
In this case the result underows.

1.9

## The truncation of mantissa leads to very interesting results. For example, 16 12 = 2

is well known. But, when the arithmetic is performed with oating point numbers,
.1667 being added 12 times yields .1996E1, whereas, .1667 12 gives .2000E1. That is,
12x = x + x +
 + x is not true.
12 times

It is very surprising that due to the truncation of mantissa, the associative and
distributive laws do not hold always in normalized oating point numbers.
That is,
(i) (a + b) + c = a + (b + c)
(ii) (a + b) c = (a c) + b
(iii) a(b c) = ab ac.
These results are illustrated in the following examples:
(i) a =.6878E1, b =.7898E1 and c =.1007E1.
Now, a + b =.1477E2
(a + b) + c = .1477E2 + .1007E1 = .1477E2 + .0100E2 = .1577E2.
Again, b + c =.8905E1.
a + (b + c)=.6878E1+.8905E1=.1578E2.
Thus, (a + b) + c = a + (b + c).
(ii) Let a =.6573E1, b =.5857E1, c =.6558E1.
Then a + b =.6631E1 and (a + b) c =.6631E1 .6558E1 = .7300E-1.
Again, a c =.1500E1 and (a c) + b =.1500E1 + .5857E1 = .7357E1.
Thus, (a + b) c = (a c) + b.
(iii) Let a =.5673E1, b =.3583E1, c =.3572E1.

## Errors in Numerical Computations

23

b c =.1100E1.
a(b c) =.5673E1 .1100E1 = .0624E0 = .6240E1.
ab =.2032E2, ac =.2026E2.
ab ac =.6000E1.
Thus, a(b c) = ab ac.
The above examples are intentionally chosen to point out the occurrence of inaccuracies in normalized oating point arithmetic due to the shifting and truncation of
numbers during arithmetic operations. But these situations always do not happen.
Here, we assume that the computer can store only four digits in mantissa, but actually
computer can store seven digits as mantissa (in single precision). The larger length of
mantissa gives more accurate result.
1.9.1

## Zeros in oating point numbers

The number zero has a denite meaning in mathematics, but, in computer exact equality
of a number to zero can never be guaranteed. The cause behind this situation is that
most of the numbers in oating point representation are approximate. One interesting
example is presented below to illustrate the behaviour of zero.

## The roots of the quadratic equation x2 + 2x 5 = 0 are x = 1 6.

The roots in oating point representation (4 digits mantissa) are .1449E1 and
.3449E1.
But, at x =.1449E1 the left hand side of the equation is .003 clearly which is not
equal to zero, while at x =.3449E1, the left hand side of the equation is
(.3449E1) (.3449E1) + .2000E1 (.3449E1) .5000E1
= .1189E2 .6898E1 .5000E1 = .1189E2 .0689E2 .0500E2 = .0000E2, which
is equal to 0.
Thus, one root satises the equation completely but other root does not, though they
are roots of the equation. By the property of the root of an equation the number 0.003
be zero. Depending on the result of this example we may note the following.
Note 1.9.1 In any computational algorithm, it is not advisable to give any instruction
based on testing whether a oating point number is zero or not.

1.10

Exercise

## 1. What do you mean by the terms in numerical analysis ?

(i) truncation error, (ii) round-o error, (iii) signicant error.
2. What are the dierent sources of computational errors in a numerical computational work ?

24 Numerical Analysis
3. Explain what do you understand by an approximate number and signicant gures
of a number.
4. What convention are used in rounding-o a number ?
5. When a number is said to be correct to n signicant gures ?
Round-o the following numbers to three signicant gures.
(i) 0.01302, (ii) 349.87, (iii) 0.005922, (iv) 87678, (v) 64.8523, (vi) 6380.7, (vii)
0.0000098, (viii) .2345, (ix) 0.4575, (x) 34.653, (xi) 21.752, (xii) 1.99999.
6. Dene absolute, relative and percentage errors.
7. Explain when relative error is a better indicator of the accuracy of a computation
than the absolute error.
8. Find out which of the following two equalities is more exact:
(i)
6/25  1.4 or 1/3  0.333, (ii) 1/9  0.1 or 1/3  0.33, (iii)  3.142 or
10  3.1623.
9. Find the absolute, relative and percentage errors when (i) 2/3 is approximated
to 0.667, (ii) 1/3 is approximated to 0.333, and (iii) true value is 0.50 and its
calculated value was 0.49.
10. (i) If is approximated as 3.14 instead of 3.14156, nd the absolute, relative and
percentage errors.
(ii) Round-o the number x = 3.4516 to three signicant gures and nd the
absolute and the relative errors.
11. The numbers 23.982 and 3.4687 are both approximate and correct only to their
last digits. Find their dierence and state how many gures in the result are
trustworthy.
12. Two lengths X and Y are measured approximately up to three signicant gures
as X = 3.32 cm and Y = 5.39 cm. Estimate the error in the computed value of
X +Y.
13. Let xT and xA denote respectively the true and approximate values of a number.
Prove that the relative error in the product xA yA is approximately equal to the
sum of the relative errors in xA and yA .
14. Show that the relative error in the product of several approximate non-zero numbers does not exceed the sum of the relative errors of the numbers.

## Errors in Numerical Computations

25

15. Show that the maximum relative error in the quotient of two approximate numbers
is approximately equal to the algebraic sum of the maximum relative errors of the
individual numbers.
16. Let x = 5.234 0.0005 and y = 5.123 0.0005. Find the percentage error of the
dierence a = x y when relative errors x = y = 0.0001.
17. What do you mean by the statement that xA (approximate value) has m signicant
gures with respect to xT (true value) ? If the rst signicant gure of xA is k
and xA is correct up to n signicant gures, prove that the relative error is less
than 101n /k.
18. Given a = 11 0.5, b = 0.04562 0.0001, c = 17200 100. Find the maximum
value of the absolute error in the following expressions
(i) a + 2b c, (ii) 2a 5b + c and (iii) a2 .
19. Calculate the quotient a = x/y of the approximate numbers x = 5.762 and y =
1.24 if all the digits of the dividend and the divisor are valid. Find the relative
and the absolute errors.
20. (i) Establish the general formula for absolute and relative errors for the function
v = f (u1 , u2 , . . . , un ) when absolute errors ui of each independent quantity ui
up uq ur
are known. Use this result for the function v = 1 s 2 t 3 to nd the upper bound
u4 u5
of the relative error.
(ii) Find the relative error in computing f (x) = 2x5 3x + 2 at x = 1, if the error
in x is 0.005.
1.42x + 3.45
(iii) If y =
here the coecients are rounded-o, nd the absolute and
x + 0.75
relative errors in y when x = 0.5 0.1.
21. Given y = x4 y 5/2 , if x0 , y0 be the approximate values of x, y respectively and
x0 , y0 be the absolute errors in them, determine the relative error in u.
(a + b)c
, where a = 1.562 0.001, b = 10.3 0.02, c = 0.12
(d e)2
0.04, d = 10.541 0.004, e = 2.34 0.006. Find the absolute and the relative
errors in the result.

22. Calculate x =

23. (i) Determine the number of correct digits in the number 0.2318 if the relative
error is 0.3 101 .
(ii) Find the number of signicant gures in the approximate number 0.4785 given
that the relative error is 0.2 102 .
24. Find the smaller root of the equation x2 500x+1 = 0 using four-digit arithmetic.

26 Numerical Analysis
25. Find the value of

103

## 26. Find an example where in an approximate computation

(i) (a + b) + c = a + (b + c), (ii) (a + b) c = (a c) + b, (iii) a(b c) = ab ac.

Chapter 2

## Calculus of Finite Dierences and

Dierence Equations
Let us consider a function y = f (x) dened on [a, b]. The variables x and y are called
independent and dependent variables respectively. The points x0 , x1 , . . . , xn are taken
as equidistance, i.e., xi = x0 + ih, i = 0, 1, 2, . . . , n. Then the value of y, when x = xi ,
is denoted by yi , where yi = f (xi ). The values of x are called arguments and that of
y are called entries. The interval h is called the dierence interval. In this chapter,
some important dierence operators, viz., forward dierence (), backward dierence
(), central dierence (), shift (E) and mean () are introduced.

2.1
2.1.1

## Finite Dierence Operators

Forward dierences

## The forward dierence or simply dierence operator is denoted by and is dened

by
f (x) = f (x + h) f (x).

(2.1)

## In terms of y, at x = xi the above equation gives

f (xi ) = f (xi + h) f (xi ), i.e., yi = yi+1 yi , i = 0, 1, 2, . . . , n 1.

(2.2)

## Explicitly, y0 = y1 y0 , y1 = y2 y1 , . . . , yn1 = yn yn1 .

The dierences of the rst dierences are called second dierences and they are
denoted by 2 y0 , 2 y1 , . . .. Similarly, one can dene third dierences, fourth dierences,
etc.
27

28 Numerical Analysis
Thus,
2 y0 = y1 y0 = (y2 y1 ) (y1 y0 ) = y2 2y1 + y0
2 y1 = y2 y1 = (y3 y2 ) (y2 y1 ) = y3 2y2 + y1
3 y0 = 2 y1 2 y0 = (y3 2y2 + y1 ) (y2 2y1 + y0 ) = y3 3y2 + 3y1 y0
3 y1 = y4 3y3 + 3y2 y1
and so on.
In general,
n+1 f (x) = [n f (x)], i.e., n+1 yi = [n yi ], n = 0, 1, 2, . . . .

(2.3)

## Also, n+1 f (x) = n [f (x + h) f (x)] = n f (x + h) n f (x)

and
n+1 yi = n yi+1 n yi , n = 0, 1, 2, . . . ,

(2.4)

## where 0 identity operator, i.e., 0 f (x) = f (x) and 1 .

The dierent forward dierences for the arguments x0 , x1 , . . . , x4 are shown in Table
2.1.
Table 2.1: Forward dierence table.
x
x0

y
y0

x1

y1

y0
2 y0
3 y0

y1
x2

2 y

y2

3 y

y2
x3

y3

x4

y4

4 y0

1
1

2 y2
y3

## Table 2.1 is called forward dierence or diagonal dierence table.

2.1.2

Backward dierences

## The backward dierence operator is denoted by and it is dened as

f (x) = f (x) f (x h).

(2.5)

29

## In terms of y, the above relation transforms to

yi = yi yi1 ,

i = n, n 1, . . . , 1.

(2.6)

That is,
y1 = y1 y0 , y2 = y2 y1 , . . . , yn = yn yn1 .

(2.7)

These dierences are called rst dierences. The second dierences are denoted by
2 y2 , 2 y3 , . . . , 2 yn . That is,
2 y2 = (y2 ) = (y2 y1 ) = y2 y1 = (y2 y1 ) (y1 y0 ) = y2 2y1 + y0 .
Similarly, 2 y3 = y3 2y2 + y1 , 2 y4 = y4 2y3 + y2 , and so on.
In general,
k yi = k1 yi k1 yi1 ,

i = n, n 1, . . . , k,

(2.8)

where 0 yi = yi , 1 yi = yi .
These backward dierences can be written in a tabular form and this table is known
as backward dierence or horizontal table.
Table 2.2 is the backward dierence table for the arguments x0 , x1 , . . . , x4 .
Table 2.2: Backward dierence table.
x
x0
x1
x2
x3
x4

2.1.3

y
y0
y1
y2
y3
y4

y1
y2
y3
y4

2 y2
2 y3
2 y4

3 y3
3 y4

4 y4

Central dierences

## The central dierence operator is denoted by and is dened by

f (x) = f (x + h/2) f (x h/2).

(2.9)

## In terms of y, the rst central dierence is

yi = yi+1/2 yi1/2

(2.10)

30 Numerical Analysis
where yi+1/2 = f (xi + h/2) and yi1/2 = f (xi h/2).
Thus y1/2 = y1 y0 , y3/2 = y2 y1 , . . . , yn1/2 = yn yn1 .
The second central dierences are
2 yi = yi+1/2 yi1/2 = (yi+1 yi ) (yi yi1 ) = yi+1 2yi + yi1 .
In general,
n yi = n1 yi+1/2 n1 yi1/2 .

(2.11)

## The central dierence table for the ve arguments x0 , x1 , . . . , x4 is shown in

Table 2.3.
Table 2.3: Central dierence table.
x
x0

y
y0

x1

y1

y1/2
2 y1
3 y3/2

y3/2
x2

2y

y2

3 y5/2

y5/2
x3

2y

y3

4 y2

y7/2
x4

y4

It is observed that all odd dierences have fraction suces and all the even dierences
are with integral suces.
2.1.4

## Shift, Average and Dierential operators

Shift operator, E:
The shift operator is dened by
Ef (x) = f (x + h).

(2.12)

Eyi = yi+1 .

(2.13)

This gives,

That is, shift operator shifts the function value yi to the next higher value yi+1 .
The second shift operator gives
E 2 f (x) = E[Ef (x)] = E[f (x + h)] = f (x + 2h).

(2.14)

## Calculus of Finite Diff. and Diff. Equs

31

In general,
E n f (x) = f (x + nh) or E n yi = yi+nh .

(2.15)

## The inverse shift operator E 1 is dened as

E 1 f (x) = f (x h).

(2.16)

## Similarly, second and higher inverse operators are

E 2 f (x) = f (x 2h)

and

E n f (x) = f (x nh).

(2.17)

## More general form of E operator is

E r f (x) = f (x + rh),

(2.18)

## where r is positive as well as negative rationals.

Average operator, :
The average operator is dened as

1
f (x + h/2) + f (x h/2)
2

1
yi = yi+1/2 + yi1/2 .
2

f (x) =
i.e.,

(2.19)

Dierential operator, D:
The dierential operator is usually denoted by D, where
d
f (x) = f  (x)
dx
d2
D2 f (x) = 2 f (x) = f  (x).
dx
Df (x) =

2.1.5

(2.20)

Factorial notation

The factorial notation has many uses in calculus of nite dierence. This is used to
nd dierent dierences and anti-dierences. The nth factorial of x, denoted by x(n) ,
is dened by
x(n) = x(x h)(x 2h) (x n 1h),
where, each factor is decreased from the earlier by h; and x(0) = 1.
Similarly, the nth negative factorial of x is dened by
1
.
x(n) =
x(x + h)(x + 2h) (x + n 1h)
It may be noted that x(n) .x(n) = 1.

(2.21)

(2.22)

32 Numerical Analysis

2.2

## Property 2.2.1 c = 0, where c is a constant.

Property 2.2.2 [f1 (x) + f2 (x) + + fn (x)]
= f1 (x) + f2 (x) + + fn (x).
Property 2.2.3 [cf (x)] = cf (x).
Combining properties (2.2.2) and (2.2.3), one can generalise the property (2.2.2) as
Property 2.2.4 [c1 f1 (x) + c2 f2 (x) + + cn fn (x)]
= c1 f1 (x) + c2 f2 (x) + + cn fn (x).
Property 2.2.5 m n f (x) = m+n f (x) = n m f (x) = k m+nk f (x),
k = 0, 1, 2, . . . , m or n.
Property 2.2.6
[f (x)g(x)] = f (x + h)g(x + h) f (x)g(x)
= f (x + h)g(x + h) f (x + h)g(x) + f (x + h)g(x) f (x)g(x)
= f (x + h)[g(x + h) g(x)] + g(x)[f (x + h) f (x)]
= f (x + h)g(x) + g(x)f (x).
Also, it can be shown that
[f (x)g(x)] = f (x)g(x) + g(x + h)f (x)
= f (x)g(x) + g(x)f (x) + f (x)g(x).

f (x)
g(x)f (x) f (x)g(x)
Property 2.2.7
=
, g(x) = 0.
g(x)
g(x + h)g(x)
Proof.

f (x + h) f (x)
f (x)
=

g(x)
g(x + h)
g(x)
f (x + h)g(x) g(x + h)f (x)
=
g(x + h)g(x)
g(x)[f (x + h) f (x)] f (x)[g(x + h) g(x)]
=
g(x + h)g(x)
g(x)f (x) f (x)g(x)
.
=
g(x + h)g(x)

33

## Property 2.2.8 In particular, when the numerator is 1, then

1
f (x)

=
.
f (x)
f (x + h)f (x)
Property 2.2.9 [cx ] = cx+h cx = cx (ch 1), for some constant c.
Property 2.2.10 [x Cr ] = xCr1 , where r is xed and h = 1.
Proof. [x Cr ] = x+1Cr xCr = xCr1 as h = 1.
Property 2.2.11 x(n) = nhx(n1) .
Proof.
x(n) = (x + h)(x + h h)(x + h 2h) (x + h n 1h)
x(x h)(x 2h) (x n 1h)
= x(x h)(x 2h) (x n 2h)[x + h {x (n 1)h}]
= nhx(n1) .
This property is analogous to the dierential formula D(xn ) = nxn1 when h = 1.
Most of the above formulae are similar to the corresponding formulae in dierential
calculus.
Property 2.2.12 The above formula can also be used to nd anti-dierence (like integration in integral calculus), as
1 x(n1) =
2.2.1

1 (n)
x .
nh

(2.23)

## Property 2.2.13 Ec = c, where c is a constant.

Property 2.2.14 E{cf (x)} = cEf (x).
Property 2.2.15 E{c1 f1 (x) + c2 f2 (x) + + cn fn (x)]
= c1 Ef1 (x) + c2 Ef2 (x) + + cn Efn (x).
Property 2.2.16 E m E n f (x) = E n E m f (x) = E m+n f (x).
Property 2.2.17 E n E n f (x) = f (x).
In particular, EE 1 I, I is the identity operator and it is some times denoted by 1.
Property 2.2.18 (E n )m f (x) = E mn f (x).

34 Numerical Analysis

Property 2.2.19 E

f (x)
g(x)

Ef (x)
.
Eg(x)

## Property 2.2.20 E{f (x) g(x)} = Ef (x) Eg(x).

Property 2.2.21 Ef (x) = Ef (x).
Property 2.2.22 m f (x) = m E m f (x) = E m m f (x)
and m f (x) = m E m f (x) = E m m f (x).

2.3

## Relations Among Operators

It is clear from the forward, backward and central dierence tables that in a denite
numerical case, the same values occur in the same positions, practically there are no
dierences among the values of the tables, but, dierent symbols have been used for the
theoretical importance.
Thus
yi = yi+1 yi = yi+1 = yi+1/2
2 yi = yi+2 2yi+1 + yi = 2 yi+2 = 2 yi+1
etc.
In general,
Again,

n yi = n yi+n ,

i = 0, 1, 2, . . . .

(2.24)

## f (x) = f (x + h) f (x) = Ef (x) f (x) = (E 1)f (x).

This relation indicates that the eect of the operator on f (x) is the same as that
of the operator E 1 on f (x). Thus
E1
Also,

or

E + 1.

(2.25)

## f (x) = f (x) f (x h) = f (x) E 1 f (x) = (1 E 1 )f (x).

That is,

1 E 1 .

(2.26)

The higher order forward dierence can be expressed in terms of the given function
values in the following way:
3 yi = (E 1)3 yi = (E 3 3E 2 + 3E 1)yi = y3 3y2 + 3y1 y0 .

## Calculus of Finite Diff. and Diff. Equs

35

There is a relation among the central dierence, , and the shift operator E, as
f (x) = f (x + h/2) f (x h/2) = E 1/2 f (x) E 1/2 f (x) = (E 1/2 E 1/2 )f (x).
That is,

E 1/2 E 1/2 .

(2.27)


1
f (x + h/2) + f (x h/2)
2
 1
1
= E 1/2 f (x) + E 1/2 f (x) = (E 1/2 + E 1/2 )f (x).
2
2
Thus,

1
(2.28)
E 1/2 + E 1/2 .
2
The average operator can also be expressed in terms of the central dierence operator.
Again,

f (x) =

2
1  1/2
E + E 1/2 f (x)
4


1
1  1/2
= (E E 1/2 )2 + 4 f (x) = 2 + 4 f (x).
4
4

2 f (x) =

Hence,

1
1 + 2.
(2.29)
4
Some more relations among the operators , , E and are deduced in the following.

## Ef (x) = f (x + h) = f (x + h) f (x) = f (x).

Also,
E 1/2 f (x) = f (x + h/2) = f (x + h) f (x) = f (x).
Thus,
E E 1/2 .
From the denition of E,
Ef (x) = f (x + h) = f (x) + hf  (x) +

h2 
h3
f (x) + f  (x) +
2!
3!

## [by Taylors series]

h3
h2
= f (x) + hDf (x) + D2 f (x) + D3 f (x) +
2!
3!

h2 2 h3 3
= 1 + hD + D + D + f (x)
2!
3!
= ehD f (x).

(2.30)

36 Numerical Analysis
Hence,
Also,

E ehD .

(2.31)

hD log E.

(2.32)

This relation is used to separate the eect of E into that of the powers of and this
method of separation is called the method of separation of symbols.
The operators and can be expressed in terms of D, as shown below

1
1
f (x) = [E 1/2 + E 1/2 ]f (x) = ehD/2 + ehD/2 f (x)
2
2
 hD 
= cosh
f (x)
2


and f (x) = [E 1/2 E 1/2 ]f (x) = ehD/2 ehD/2 f (x)
 hD 
= 2 sinh
f (x).
2
Thus,
cosh

 hD 

Again,
2 cosh

2
 hD 
2

and 2 sinh

sinh

 hD 
2

 hD 
2

= sinh(hD).

(2.33)

(2.34)

## The inverse relation

hD sinh1 ()

(2.35)

is also useful.
Since E 1 + and E 1 1 , [from (2.25) and (2.26)]
from (2.32), it is obtained that
hD log E log(1 + ) log(1 ) sinh1 ().

(2.36)

## The operators and E are commutative, as

Ef (x) = f (x + h) =


1
f (x + 3h/2) + f (x + h/2) ,
2

while
Ef (x) = E
Hence,

 1 

f (x + h/2) + f (x h/2) = f (x + 3h/2) + f (x + h/2) .
2
2

1

E E.

(2.37)

37

## Example 2.3.1 Prove the following relations.



2
2 2

2
2 2
1/2
+ 1+ ,
, (ii) E
+ , (iii)
(i) 1 + 1 +
2
2
2
4
E 1
+
(iv) (1 + )(1 ) 1, (v)
+ , (vi)
,
2
2
2
(vii) 2 .
Solution. (i) f (x) = 12 (E 1/2 + E 1/2 )(E 1/2 E 1/2 )f (x) = 12 [E E 1 ]f (x).
Therefore,

1
2 2
1 2
(1 + )f (x) = 1 + (E E ) f (x)
4

1 2
1
= 1 + (E 2 + E 2 ) f (x) = (E + E 1 )2 f (x)
4
4

2 2
1 1/2
1/2 2
) f (x) = 1 +
f (x).
= 1 + (E E
2
2


2 2
.
1+ 1+
2

Hence

2 2

(2.38)





1 1/2

1 1/2
1/2
1/2
f (x) =
[E + E
(ii) +
] + [E E
] f (x) = E 1/2 f (x).
2
2
2
Thus

E 1/2 + .
2
(iii)

2
+
2

(2.39)

2
f (x)
1+
4

1 1/2
1
1/2 2
1/2
1/2
= (E E
) f (x) + (E E
) 1 + (E 1/2 E 1/2 )2 f (x)
2
4
1
1
= [E + E 1 2]f (x) + (E 1/2 E 1/2 )(E 1/2 + E 1/2 )f (x)
2
2
1
1
= [E + E 1 2]f (x) + (E E 1 )f (x)
2
2
= (E 1)f (x).
Hence,

2
+
2

1+

2
E 1 .
4

(2.40)

38 Numerical Analysis
(iv) (1 + )(1 )f (x) = (1 + )[f (x) f (x) + f (x h)]
= (1 + )f (x h) = f (x h) + f (x) f (x h)
= f (x).
Therefore,
(1 + )(1 ) 1.

(v)

(2.41)

E 1
1
+
f (x) = [f (x h) + f (x)]
2
2
2

1
= [f (x) f (x h) + f (x + h) f (x)]
2
1
1
= [f (x + h) f (x h)] = [E E 1 ]f (x)
2
2
1 1/2
= (E + E 1/2 )(E 1/2 E 1/2 )f (x)
2
= f (x).
Hence
E 1
+
.
2
2
(vi)

(2.42)

+
1
f (x) = [f (x) + f (x)]
2
2
1
= [f (x + h) f (x) + f (x) f (x h)]
2
1
1
= [f (x + h) f (x h)] = [E E 1 ]f (x)
2
2
= f (x)
(as in previous case).

Thus,

+
.
2

(2.43)

## (vii) f (x) = [f (x) f (x h)] = f (x + h) 2f (x) + f (x h).

Again,
f (x) = f (x + h) 2f (x) + f (x h) = (E 2 + E 1 )f (x)
= (E 1/2 E 1/2 )2 f (x) = 2 f (x).
Hence,

(E 1/2 E 1/2 )2 2 .

(2.44)

## Calculus of Finite Diff. and Diff. Equs

39

The relations among the various operators are shown in Table 2.4.
Table 2.4: Relationship between the operators.
E

+1

(1 )1

E1

(1 )1 1

1 E 1

1 (1 + )1

## E 1/2E 1/2 (1 + )1/2 (1 )1/2

E 1/2+E 1/2
(1 + /2) (1/2)(1)1/2

2
(1 + )1/2
hD
log E
log(1 + )
log(1 )

hD
2

+ 1+
ehD
4

2
2
+ 1+
ehD 1
2
4

2
2
1 ehD
+ 1+
2
4

2 sinh(hD/2)
2

cosh(hD/2)
1+
4
1+

2 sinh1 (/2)

hD

## From denition of derivative, we have

f (x + h) f (x)
f (x)
= lim
.
h0
h0
h
h

f  (x) = lim
Thus one can write,

f (x)  hf  (x).

Again,
f  (x + h) f  (x)
h0
h
f (x + h) f (x)

h
h
 lim
h0
h
f (x + h) f (x)
2 f (x)
= lim
=
lim
.
h0
h0
h2
h2

f  (x) = lim

## Therefore, h2 f  (x)  2 f (x).

In general, n f (x)  hn f n (x). That is, for small h, the operators and hD are
almost equal.

2.4

## From the denition of factorial notation,

40 Numerical Analysis

x(0)
x(1)
x(2)
x(3)
x(4)

=1
=x
= x(x h)
= x(x h)(x 2h)
= x(x h)(x 2h)(x 3h)

(2.45)

and so on.
The above relations show that x(n) , n = 1, 2, . . . is a polynomial of degree n in x.
Also, x, x2 , x3 , . . . can be expressed in terms of factorial notations x(1) , x(2) , x(3) , . . ., as
shown below.
1 = x(0)
x = x(1)
x2 = x(2) + hx(1)
x3 = x(3) + 3hx(2) + h2 x(1)
x4 = x(4) + 6hx(3) + 7h2 x(2) + h3 x(1)

(2.46)

and so on.
These relations show that xn can be expressed as a polynomial of x(1) , x(2) , . . . , x(n) ,
of degree n. Once a polynomial is expressed in a factorial notation, its dierences can
be obtained by using the formula like dierential calculus.
Example 2.4.1 Express f (x) = 2x4 + x3 5x2 + 8 in factorial notation and nd
its rst and second dierences.
Solution. Here we assume that h = 1.
Then by (2.46), x = x(1) , x2 = x(2) + x(1) , x3 = x(3) + 3x(2) + x(1) ,
x4 = x(4) + 6x(3) + 7x(2) + x(1) .
Using these values, the function f (x) becomes

 



f (x) = 2 x(4) + 6x(3) + 7x(2) + x(1) + x(3) + 3x(2) + x(1) 5 x(2) + x(1) + 8
= 2x(4) + 13x(3) + 12x(2) 2x(1) + 8.
Now, the Property 2.2.11, i.e., x(n) = nx(n1) is used to nd the dierences.
Therefore,
f (x) = 2.4x(3) + 13.3x(2) + 12.2x(1) 2.1x(0) = 8x(3) + 39x(2) + 24x(1) 2
and 2 f (x) = 24x(2) + 78x(1) + 24.
In terms of x,
f (x) = 8x(x 1)(x 2) + 39x(x 1) + 24x 2
and 2 f (x) = 24x(x 1) + 78x + 24.
From the relations of (2.46) one can conclude the following result.

## Calculus of Finite Diff. and Diff. Equs

41

Lemma 2.4.1 Any polynomial f (x) in x of degree n can be expressed in factorial notation with same degree, n.
This means, in conversion to the factorial notation, the degree of a polynomial remains
unchanged.
The above process to convert a polynomial in a factorial form is a labourious technique when the degree of the polynomial is large. The other systematic process, like
Maclaurins formula in dierential calculus, is used to convert a polynomial, even a
function, in factorial notation.
Let f (x) be a polynomial in x of degree n. In factorial notation, let it be
f (x) = a0 + a1 x(1) + a2 x(2) + + an x(n) ,

(2.47)

## where ai s are unknown constants to be determined, an = 0.

Now, one can nd the dierences of (2.47) as follows.
f (x) = a1 + 2a2 x(1) + 3a3 x(2) + + nan x(n1)
2 f (x) = 2.1a2 + 3.2a3 x(1) + + n(n 1)an x(n2)
3 f (x) = 3.2.1a3 + 4.3.2.x(1) + + n(n 1)(n 2)an x(n3)

n f (x) = n(n 1)(n 2) 3 2 1an = n!an .
When x = 0, the above relations give
f (0) = a1 ,
a0 = f (0),
2 f (0)
2 f (0) = 2.1.a2 or, a2 =
2!
3 f (0)
3
f (0) = 3.2.1.a3 or, a3 =
3!

n f (0)
n f (0) = n!an
or, an =
.
n!
Using these results equation (2.47) becomes
f (x) = f (0) + f (0)x(1) +

n f (0) (n)
x +
x + +
x .
2!
3!
n!
(2.48)

## This formula is similar to Maclaurins formula of dierential calculus and it is also

used to expand a function in terms of factorial notation. To perform the formula (2.48),
dierent forward dierences are to be evaluated at x = 0, and this can be done using
forward dierence table. This is a systematic process and easy to implement as computer
program.

42 Numerical Analysis

## Example 2.4.2 Express f (x) = 2x4 5x3 + 8x2 + 2x 1 in factorial notation.

Solution. Taking h = 1. f (0) = 1, f (1) = 6, f (2) = 27, f (3) = 104, f (4) = 327.
x
0

f (x)
1

f (x)

2 f (x)

3 f (x)

4 f (x)

7
14
21
2

27

104

327

42
56

48

77

90
146

223
Thus using formula (2.48)

## 2 f (0) (2) 3 f (0) (3) 4 f (0) (4)

x +
x +
x
2!
3!
4!
= 2x(4) + 7x(3) + 7x(2) + 7x(1) 1.

## Example 2.4.3 If f (x) = x4 + 2x3 + 8x2 + 3, nd f (x).

Solution. Applying synthetic division to express f (x) in factorial notation.
1

1
1

2
1
3
2
5
3
8

8
3
11
10
21

0
11
11

## Therefore, f (x) = x(4) + 8x(3) + 21x(2) + 11x(1) + 3.

Hence,
8
21
11
1
f (x) = x(5) + x(4) + x(3) + x(2) + 3x(1) + c, [using Property 2.2.12]
5
4
3
2
1
= x(x 1)(x 2)(x 3)(x 4) + 2x(x 1)(x 2)(x 3)
5
11
+7x(x 1)(x 2) + x(x 1) + 3x + c, where c is arbitrary constant.
2

## Calculus of Finite Diff. and Diff. Equs

2.5

43

Dierence of a Polynomial

## Let f (x) = a0 xn + a1 xn1 + + an1 x + an be a polynomial of degree n. From this

expression one can nd the successive dierences of f (x).
Now, f (x) = f (x + h) f (x), where h is the spacing of x
= a0 [(x + h)n xn ] + a1 [(x + h)n1 xn1 ] + + an1 [(x + h) x].
Expanding the terms within parenthesis using binomial theorem and obtain


n(n 1) 2 n2
f (x) = a0 xn + nhxn1 +
h x
+ + hn xn
2!


(n 1)(n 2) 2 n3
h x
+a1 xn1 + (n 1)hxn2 +
+ + hn1 xn1
2!
+ + han1
 n(n 1)

= a0 nhxn1 + a0
h2 + a1 (n 1)h xn2 + + an1 h.
2!
If h is constant, then the coecients of xn1 , xn2 , . . . , x and an1 h are constants.
The coecients of xn1 , xn2 , . . . , x and the constant term are denoted by b0 , b1 , . . . , bn2
and bn1 respectively. In this notation rst dierence can be written as
f (x) = b0 xn1 + b1 xn2 + + bn2 x + bn1 , where b0 = a0 nh.
It may be noted that f (x) is a polynomial of degree n 1, i.e., rst dierence
reduces the degree of f (x) by 1.
The second dierence of f (x) is
2 f (x)
= f (x + h) f (x)
= b0 [(x + h)n1 xn1 ] + b1 [(x + h)n2 xn2 ] + + bn2 [(x + h) x]


(n 1)(n 2) 2 n3
h x
+ + hn1 xn1
= b0 xn1 + (n 1)hxn2 +
2!


(n

2)(n 3) 2 n4
h x
+b1 xn2 + (n 2)hxn3 +
+ + hn2 xn2
2!
+ + bn2 h
1

= b0 (n 1)hxn2 +
(n 1)(n 2)b0 h2 + (n 2)b1 h xn3 + + bn2 h
2!
= c0 xn2 + c1 xn3 + + cn3 x + cn2
where c0 = b0 (n 1)h, c1 = 2!1 (n 1)(n 2)b0 h2 + (n 2)b1 h, etc.
This expression shows that 2 f (x) is a polynomial of degree n 2.
It may be noted that the coecient of the leading term is c0 = b0 h(n 1) = n(n
1)h2 a0 and it is a constant quantity.

44 Numerical Analysis
In this way, one can nd n1 f (x) is a polynomial of degree one and let it be p0 x+p1 ,
i.e., n1 f (x) = p0 x + p1 .
Then n f (x) = p0 (x+h)+p1 p0 xp1 = p0 h, which is a constant. And n+1 f (x) =
0.
It can be shown that n f (x) = n(n 1)(n 2) 2 1 hn a0 = n!hn a0 .
Thus nally,
k f (x), k < n is a polynomial of degree n k,
n f (x) is constant, and
k f (x), k > n is zero.

Alternative proof.
It is observed that, a polynomial in x of degree n can be expressed as a polynomial in
factorial notation with same degree n.
Thus, if f (x) = a0 xn + a1 xn1 + a2 xn2 + + an1 x + an be the given polynomial
then it can be written as f (x) = b0 x(n) + b1 x(n1) + b2 x(n2) + + bn1 x(1) + bn .
Therefore,
f (x) = b0 nhx(n1) + b1 h(n 1)x(n2) + b2 h(n 2)x(n3) + + bn1 h.
Clearly this is a polynomial of degree n 1.
Similarly,
2 f (x) = b0 n(n 1)h2 x(n2) + b1 (n 1)(n 2)h2 x(n3) + + bn2 h2 ,
3 f (x) = b0 n(n 1)(n 2)h3 x(n3) + b1 (n 1)(n 2)(n 3)h3 x(n4)
+ + bn3 h3 .
In this way,
n f (x) = b0 n(n 1)(n 2) 2 1 hn x(nn)
= b0 n!hn ,

a constant quantity.

## Hence n+1 f (x) = 0.

Dierence of factorial power function
From denition of factorial notation, we have
x(n) = nhx(n1)
2 x(n) = nhx(n1) = nh.(n 1)hx(n2) = n(n 1)h2 x(n2)
3 x(n) = n(n 1)h2 .(n 2)hx(n2) = n(n 1)(n 2)h3 x(n3) .

## Calculus of Finite Diff. and Diff. Equs

45

In this way,
n x(n) = n(n 1)(n 2) 2 1 hn x(nn) = n!h2 .
Example 2.5.1 Given xi = x0 + ih, i = 0, 1, 2, . . . , n; h > 0
and ui (x) = (x x0 )(x x1 ) (x xi ),
prove that
k ui (x) = (i + 1)i(i 1) (i k + 2)hk (x x0 )(x x1 ) (x xik ).
Solution. Here ui (x) = (x x0 )(x x1 ) (x xi ) = (x x0 )(i+1) (say).
Therefore,
ui (x) = (x + h x0 )(x + h x1 ) (x + h xi ) (x x0 ) (x xi )
= (x + h x0 )(x x0 )(x x1 ) (x xi1 )
(x x0 )(x x1 ) (x xi )
= (x x0 )(x x1 ) (x xi1 )[(x + h x0 ) (x xi )]
= (x x0 )(x x1 ) (x xi1 )(h + xi x0 )
[since xi = x0 + ih]
= (x x0 )(x x1 ) (x xi1 )(i + 1)h
= (i + 1)h(x x0 )(i) .
Similarly,
2 ui (x) = (i + 1)h[(x + h x0 )(x + h x1 ) (x + h xi1 )
(x x0 )(x x1 ) (x xi1 )]
= (i + 1)h(x x0 )(x x1 ) (x xi2 )[(x + h x0 ) (x xi1 )]
= (i + 1)h(x x0 )(i1) ih
= (i + 1)ih2 (x x0 )(i1) .
In similar way,
3 ui (x) = (i + 1)i(i 1)h3 (x x0 )(i2) .
Hence,
k ui (x) = (i + 1)i(i 1) (i k 2)hk (x x0 )(ik1)
= (i + 1)i(i 1) (i k + 2)hk (x x0 )(x x1 ) (x xik ).

2.6

Summation of Series

The nite dierence method is also used to nd the sum of a nite series. Two important
results are presented here.

46 Numerical Analysis
Theorem 2.1 If f (x) be dened only for integral values of independent variable x, then

b+h
b

f (x) = f (a) + f (a + h) + f (a + 2h) + + f (b) = F (x)
i.e.,

x=a
b


f (x) = F (x + h) F (x),

b = a + nh for some n

(2.49)

x=a

## where F (x) is an anti-dierence (instead of anti-derivative) of f (x),

i.e., F (x) = f (x).
Proof. Since F (x) = f (x), therefore,
b

x=a

f (x) =

b


F (x) =

x=a

b


[F (x + h) F (x)]

x=a

## = [F (b + h) F (b)] + [F (b) F (b h)] + + [F (a + h) F (a)]

= F (b + h) F (a).
Thus, if F (x) is anti-dierence of f (x), i.e.,
b

1 f (x) = F (x), then
f (x) = F (b + h) F (a).
x=a

Example 2.6.1 Use nite dierence method to nd the sum of the series
n

3
.
f (x), where f (x) = x(x 1) +
x(x + 1)(x + 2)
x=1

Solution. Let

F (x) =

f (x) =

x(x 1) +

3
x(x + 1)(x + 2)

x(2)
x(3)
+3
3
2
3
1
1
.
= x(x 1)(x 2)
3
2 x(x + 1)

= 1 x(2) + 31 x(3) =

Therefore,
n

x=1

3
1
3 1
1
0+
f (x) = [F (n + 1) F (1)] = (n + 1)n(n 1)
3
2 (n + 1)(n + 2)
2 1.2
1
1
3
3
= n(n2 1)
+ .
3
2 (n + 1)(n + 2) 4

## Calculus of Finite Diff. and Diff. Equs

47

Summation by parts
Like the formula integration by parts of integral calculus there is a similar formula in
nite dierence calculus. If f (x) and g(x) are two functions dened only for integral
values of x between a and b, then
b


b

b+h 
f (x)g(x) = f (x)g(x)

a

x=a

x=a
n


x4x .

x=1

## Solution. Let f (x) = x, g(x) = 4x . Then g(x) = 4x /3. Hence for h = 1,

n
n

 4x n+1 
4n+1 4  4  x
4x+1
.x = (n + 1)

x4x = x.

4 .1
3 1
3
3
3
3
x=1
x=1
x=1

4n+1
4n+1 4  4 4(4n 1)

=n
.
= (n + 1)
3
3
3
4
3

n


2.7

## Example 2.7.1 If n is a positive integer then show that

1
, where h is the spacing.
x(n) =
(x + nh)(n)
Solution. From the denition of x(n) , we have,
x(n) = x(x h)(x 2h) (x n 2h)(x n 1h) = x(n1) (x n 1h).
x(n)
.
This can be written as x(n1) =
x (n 1)h
Substituting n = 0, 1, 2, . . . , (n 1) we obtain
1
x(0)
=
x+h
x+h
x(1)
1
=
=
x + 2h
(x + h)(x + 2h)

x(1) =
x(2)

x(3) =

1
x(2)
=
.
x + 3h
(x + h)(x + 2h)(x + 3h)

(2.50)

48 Numerical Analysis
In this way,
x(n1) =

1
(x + h)(x + 2h) (x + n 1h)

1
(x + nh)(x + nh h)(x + nh 2h) (x + nh n 1h)
1
.
=
(x + nh)(n)

## Example 2.7.2 Prove the following identities

(i) u0 + u1 + u2 + + un = n+1 C1 u0 + n+1 C2 u0 + n+1 C3 2 u0 + + n u0 .


x2
x3
x2
(ii) u0 + xu1 + u2 + u3 + = ex u0 + xu0 + 2 u0 +
2!
3!
2!
1
1
1 2
1
(iii) u0 u1 + u2 u3 + = u0 u0 + u0 3 u0 + .
2
4
8
16
Solution. (i) u0 + u1 + u2 + + un
= u0 + Eu0 + E 2 u0 + + E n u0

 n+1
E
1
2
n
u0
= (1 + E + E + + E )u0 =
E1


(1 + )n+1 1
u0
=
[since E 1 + ]


1  n+1
C1 + n+1 C2 2 + + n+1 u0
=

## = n+1 C1 u0 + n+1 C2 u0 + n+1 C3 2 u0 + + n u0 .

(ii)

x2
x3
u2 + u3 +
2!
3!
x2 2
x3
= u0 + xEu0 + E u0 + E 3 u0 +
2!
3!

2
(xE)3
(xE)
+
+ u0
= 1 + xE +
2!
3!
u0 + xu1 +

= exE u0 = ex(1+) u0 = ex ex u0

(x)2 (x)3
+
+ u0
= ex 1 + x +
2!
3!

2
x 2
x3 3
x
= e u0 + xu0 + u0 + u0 +
2!
3!

## Calculus of Finite Diff. and Diff. Equs

(iii)
u0 u 1 + u2 u 3 +
= u0 Eu0 + E 2 u0 E 3 u0 + = [1 E + E 2 E 3 + ]u0
1
1
1+
= (1 + E)1 u0 = (2 + )1 u0 =
u0
2
2

1
1
u0 2 u0 3 u0
2 3

+ u0 = u0
+

+
=
1 +
2
2
4
8
2
4
8
16

i

fi = E f0 =

i  

i
j=0

j f0 .

## Solution. From denition of ,

f (xi ) = f (xi + h) f (xi ) = Ef (xi ) f (xi ),
i.e., Ef (xi ) = f (xi ) + f (xi ) = ( + 1)f (xi ).
Hence, E + 1 and also E i (1 + )i .
Therefore,
fi = E f0 = (1 + ) f0 = 1 + C1 f0 + C2 f0 + =
i

i  

i
j=0

j f0 .

## Example 2.7.4 Prove that

n f (x) =

n


(1)i n Ci f [x + (n i)h],

i=0

where h is step-length.
Solution. We know that, E 1 and n (E 1)n .
n

(1)i n Ci E ni f (x).
Therefore, n f (x) = (E 1)n f (x) =
i=0

## [by Binomial theorem]

Now, Ef (x) = f (x + h), E 2 f (x) = f (x + 2h), . . . , E ni f (x) = f [x + (n i)h].
n

n
(1)i n Ci f [x + (n i)h].
Hence f (x) =
i=0

49

50 Numerical Analysis

Example 2.7.5 Find the polynomial f (x) which satises the following data and
hence nd the value of f (1.5).
x
f (x)

:
:

1
3

2
5

3
10

4
30

x
1

f (x)
3

f (x)

2 f (x)

3 f (x)

2
3
5
3

12

10

15
20

30

## It is known that, f (x0 + nh) = E n f (x0 ) = (1 + )n f (x0 ).

Here x0 = 1, h = 1. Let x0 + nh = 1 + n = x, i.e., n = x 1. Therefore,
f (x) = E x1 f (x0 ) = (1 + )x1 f (x0 )
(x 1)(x 2) 2
f (0)
= f (0) + (x 1)f (0) +
2!
(x 1)(x 2)(x 3) 3
f (0) +
+
3!
3(x 1)(x 2) 12(x 1)(x 2)(x 3)
+
= 3 + 2(x 1) +
2!
3!
21 2 39
3
= 2x x + x 8.
2
2
Thus f (1.5) = 4.375.
Example 2.7.6 Find the missing term in the following table:
x
f (x)

:
:

1
2

2
3

3
8

5
21

Solution. Here four values of f (x) are given. So, we consider f (x) be a polynomial
of degree 3. Thus the fourth dierences of f (x) vanish, i.e.,
4 f (x) = 0 or, (E 1)4 f (x) = 0
or, (E 4 4E 3 + 6E 2 4E + 1)f (x) = 0
or, E 4 f (x) 4E 3 f (x) + 6E 2 f (x) 4Ef (x) + f (x) = 0

## Calculus of Finite Diff. and Diff. Equs

51

or, f (x + 4) 4f (x + 3) + 6f (x + 2) 4f (x + 1) + f (x) = 0.
Here, h = 1 as the values are in spacing of 1 unit.
For x = 1 the above equation becomes
f (5) 4f (4) + 6f (3) 4f (2) + f (1) = 0 or, 21 4f (4) + 6 8 4 3 2 = 0
or, f (4) = 13.75.
Example 2.7.7 Use nite dierence method to nd the values of a and b in the
following table.
x
f (x)

:
:

0
5

2
a

4
8

6
b

8
20

10
32

Solution. Here, four values of f (x) are known, so we can assume that f (x) is a
polynomial of degree 3. Then, 4 f (x) = 0.
or, (E 1)4 f (x) = 0
or, E 4 f (x) 4E 3 f (x) + 6E 2 f (x) 4Ef (x) + f (x) = 0
or, f (x + 8) 4f (x + 6) + 6f (x + 4) 4f (x + 2) + f (x) = 0
[Here h = 2, because the values of x are given in 2 unit interval]
In this problem, two unknowns a and b are to be determined and needs two equations.
Therefore, the following equations are obtained by substituting x = 2 and x = 0 to
the above equation.
f (10) 4f (8) + 6f (6) 4f (4) + f (2) = 0 and
f (8) 4f (6) + 6f (4) 4f (2) + f (0) = 0.
These equations are simplies to
32 4 20 + 6b 4 8 + a = 0 and 20 4b + 6 8 4a 5 = 0.
That is, 6b + a 80 = 0 and 4b 4a + 63 = 0. Solution of these equations is
a = 2.9, b = 12.85.

Example 2.7.8 Find the value of


2 2
x .
E

Solution.


(E 1)2 2
2 2
x =
x
E
E

2
E 2E + 1 2
x
=
E


## = (E 2 + E 1 )x2 = Ex2 2x2 + E 1 x2

= (x + 1)2 2x2 + (x 1)2
= 2.

52 Numerical Analysis

f (x)
.
Example 2.7.9 Show that log f (x) = log 1
f (x)

Solution.
log f (x) = log f (x) log f (x h) = log f (x) log E 1 f (x)
E 1 f (x)
(1 )f (x)
f (x)
= log
= log
= log 1
E f (x)
f (x)
f (x)

f (x)
f (x) f (x)
= log 1
.
= log
f (x)
f (x)
The following formulae for anti-dierence can easily be veried, for h = 1.
(i) 1 cf (x) = c1 f (x), c being a constant.
(ii) 1 x(n) =
(iii) 1 ax =

1
(n+1) , n
n+1 x

1
ax ,
a1

(iv) 1 sin ax =
(v) 1 cos ax =

2.8

## being a positive integer.

a = 1.

1
cos(ax a/2).
2 sin a/2

1
sin(ax a/2).
2 sin a/2

Dierence Equations

## Like dierential equations, dierence equations have many applications in dierent

branches of mathematics, statistics and other eld of science and engineering.
A dierence equation is an equation containing an independent variable, a dependent
variable and successive dierences of dependent variable. The dierence equations are
some times called recurrence equations.
For example,
a2 f (x) + bf (x) + cf (x) = g(x),

(2.51)

where a, b, c are constants, g(x) is a known function and f (x) is the unknown function.
The solution of a dierence equation is the value of the unknown function.
The dierence equation can also be expressed as a relation among the independent
variable and the successive values, i.e., f (x), f (x + h), f (x + 2h), . . . , of dependent
variable. For example, the dierence equation
22 f (x) f (x) + 5f (x) = x2 + 3x,

(2.52)

53

## is same as 2E 2 f (x) 5Ef (x) + 8f (x) = x2 + 3x,

or, it can be written as
2f (x + 2h) 5f (x + h) + 8f (x) = x2 + 3x.
If f (x) is denoted by ux , then this equation can be written as
2ux+2h 5ux+h + 8ux = x2 + 3x.
When h = 1, the above equation is simplied as
2ux+2 5ux+1 + 8ux = x2 + 3x.

(2.53)

The dierence between the largest and the smallest arguments appearing in the difference equation with unit interval is called the order of the dierence equation.
The order of the equation (2.53) is (x + 2) x = 2, while the order of the equation
ux+3 8ux+1 + 5ux1 = x3 + 2 is (x + 3) (x 1) = 4. The order of the dierence
equation f (x + 2) 3f (x) = 0 is 3 as it is equivalent to ux+3 ux+2 3ux = 0.
A dierence equation in which ux , ux+1 , . . . , ux+n occur to the rst degree only and
there are no product terms is called linear dierence equation. Its general form is
a0 ux+n + a1 ux+n1 + + an ux = g(x).

(2.54)

If the coecients a0 , a1 , . . . , an are constants, then the equation (2.54) is called linear
dierence equation with constant coecients. If g(x) = 0 then the equation is
called homogenous otherwise it is called non-homogeneous dierence equation.
The linear homogeneous equation with constant coecients of order n is
a0 ux+n + a1 ux+n1 + + an ux = 0.

(2.55)

## This can be written as

(a0 E n + a1 E n1 + + an )ux = 0

or,

f (E)ux = 0,

(2.56)

## where f (E) a0 E n + a1 E n1 + + an is known as the characteristic function of

(2.56).
The equation f (m) = 0 is called the auxiliary equation (A.E.) for the dierence
equation (2.56).
The solution of a dierence equation is a relation between the independent variable
and the dependent variable satisfying the equation.
2.8.1

## Let {1, 2, 4, 8, 16, . . .} be a sequence and its general term is 2n , n = 0, 1, 2, . . .. Let us

denote the nth term by an . Then an = 2n . Also, an+1 = 2n+1 . Thus an+1 = 2.2n = 2an ,
i.e., an+1 2an = 0 is the dierence equation of the above sequence.

54 Numerical Analysis
Let us consider another example of a sequence whose xth term is given by
ux = a2x + b3x

(2.57)

(2.58)

(2.59)

Then,

and

## Now ux+1 2ux = b3x [using (2.57) and (2.58)]

and ux+2 2ux+1 = b3x+1 . [using (2.58) and (2.59)]
Therefore,
ux+2 2ux+1 = 3(b3x ) = 3(ux+1 2ux ).
That is, ux+2 5ux+1 + 6ux = 0 is the required dierence equation for the relation
(2.57).
It may be noted that the equation (2.57) contains two arbitrary constants and the
corresponding dierence equation is of order 2.
A large number of counting problems can be modelled by using recurrence relations.
Some of them are presented here.
Example 2.8.1 (Rabbits on an island). This problem was originally posed by
the Italian mathematician Leonardo Fibonacci in the thirteenth century in his book
Liber abaci. The problem is stated below.
A pair of new-born rabbits (one male and other female) is kept on an island where
there is no other rabbit. A pair of rabbits does not breed until they are 2 months old.
After a pair becomes 2 months old, each pair of rabbits (of opposite sexes) produces
another pair (of opposite sexes) each month. It is assuming that no rabbits ever die.
This problem can be modelled as a recurrence relation as follows.
Let xn denote the number of pairs of rabbits on the island just after n months. At
the end of rst month the number of pairs of rabbits is 1, i.e., x1 = 1. Since this pair
does not breed during second month, so x2 = 1. Now,
xn = number of pairs of rabbits just after n months
= number of pairs after (n 1) months + number of new born pairs at
the end of nth month.
The number of new born pairs = number of pairs just after the (n 2)th month,
since each new-born pair is produced by a pair of at least 2 months old, and it is
equal to xn2 .

## Calculus of Finite Diff. and Diff. Equs

55

Hence
xn = xn1 + xn2 ,

n 3,

x1 = x2 = 1.

(2.60)

This is the dierence equation of the above stated problem and the solution is x1 =
1, x2 = 1, x2 = 2, x3 = 3, x4 = 5, x5 = 8, . . ., i.e., the sequence is {1, 1, 2, 3, 5, 8 . . .}
and this sequence is known as Fibonacci sequence.
Example 2.8.2 (The Tower of Hanoi). The Tower of Hanoi problem is a famous
problem of the late nineteenth century. The problem is stated below.
Let there be three pegs, numbered 1, 2, 3 and they are on a board and n discs of
dierence sizes with holes in their centres. Initially, these n discs are placed on one
peg, say peg 1, in order of decreasing size, with the largest disc at the bottom. The
rules of the puzzle are that the discs can be moved from one peg to another only one
at a time and no discs can be placed on the top of a smaller disc. The problem is to
transfer all the discs from peg 1 to another peg 2, in order of size, with the largest
disc at the bottom, in minimum number of moves.
Let xn be the number of moves required to solve the problem with n discs. If n = 1,
i.e., if there is only one disc on peg 1, we simply transfer it to peg 2 by one move.
Hence x1 = 1. Now, if n > 1, starting with n discs on peg 1 we can transfer the top
n 1 discs, following the rules of this problem to peg 3 by xn1 moves. During these
moves the largest disc at the bottom on peg 1 remains xed. Next, we use one move
to transfer the largest disc from peg 1 to peg 2, which was empty. Finally, we again
transfer the n 1 discs on peg 3 to peg 2 by xn1 moves, placing them on top of the
largest disc on peg 2 which remains xed during these moves. Thus, when n > 1,
(n 1) discs are transferred twice and one additional move is needed to move the
largest disc at the bottom from peg 1 to peg 2. Thus the recurrence relation is
xn = 2xn1 + 1 for n 2 and x1 = 1.

(2.61)

2.9

## Solution of Dierence Equations

Several methods are used to solve dierence equations. Among them the widely used
methods are iterative method, solution using operators, solution using generating function, etc.
2.9.1

Iterative method

In this method the successive terms are substituted until the terms reduce to initial
term. The method is illustrated by example.

56 Numerical Analysis

## Example 2.9.1 Solve the dierence equation

xn = xn1 + (n 1) for n 2 and x1 = 0.
Solution.

## xn = xn1 + (n 1) = {xn2 + (n 2)} + (n 1)

= xn2 + 2n (1 + 2) = {xn3 + (n 3)} + 2n (1 + 2)
= xn3 + 3n (1 + 2 + 3).

## In this way, after (n 1) steps

xn = xn(n1) + (n 1)n (1 + 2 + + n 1)
= x1 + (n 1)n

n(n 1)
2

1
= 0 + n(n 1) [since x1 = 0]
2
1
= n(n 1).
2
Example 2.9.2 Solve the dierence equation for the Tower of Hanoi problem: xn =
2xn1 + 1, n 2 with x1 = 1.
Solution.

xn = 2xn1 + 1 = 2(2xn2 + 1) + 1
= 22 xn2 + (2 + 1) = 22 (2xn3 + 1) + (2 + 1)
= 23 xn3 + (22 + 2 + 1) = 23 (2xn4 + 1) + (22 + 2 + 1)
= 24 xn4 + (23 + 22 + 2 + 1).

## In this way, after (n 1) steps,

xn = 2n1 xn(n1) + (2n2 + 2n3 + + 22 + 2 + 1)
= 2n1 x1 + (2n2 + 2n3 + + 22 + 2 + 1)
[since x1 = 1]
= 2n1 + 2n2 + 2n3 + + 22 + 2 + 1
n
= 2 1.

2.9.2

## Solution using symbolic operators

This method is used to solve homogeneous as well as non-homogeneous dierence equations. First we consider the homogeneous linear dierence equations with constant
coecients.

57

## Solution of homogeneous equations with constant coecients

To explain the method, a general second order dierence equation is considered. Let
ux+2 + aux+1 + bux = 0.

(2.62)

## Using shift operator E, this equation can be written as

(E 2 + aE + b)ux = 0.

(2.63)

## Let ux = cmx be a solution of (2.63), c is a non-zero constant.

Then, E 2 ux = ux+2 = cmx+2 , Eux = ux+1 = cmx+1 .
Using these values, the equation (2.63) reduces to cmx (m2 + am + b) = 0. Since
cmx = 0,
m2 + am + b = 0.

(2.64)

This equation is called the auxiliary equation (A.E.) for the dierence equation
(2.62). Since (2.64) is a quadratic equation, three types of roots may occur.
Case I. Let m1 and m2 be two distinct real roots of (2.64). In this case, the general
solution is ux = c1 mx1 + c2 mx2 , where c1 and c2 are arbitrary constants.
Case II. Let m1 , m1 be two real and equal roots of (2.64). In this case (c1 mx1 + c2 mx1 ) =
(c1 + c2 )mx1 = cmx1 is the only one solution of (2.62). To get the other solution (as a
second order dierence equation should have two independent solutions, like dierential
equation), let us consider ux = mx1 vx be its solution.
Since m1 , m1 are two equal roots of (2.64), the equation (2.63) may be written as
(E 2 2m1 E + m21 )ux = 0.
Substituting ux = mx1 vx to this equation, we obtain
x+2
x+2
mx+2
1 ux+2 2m1 vx+1 + m1 vx = 0
x+2
2
or, m1 (vx+2 2vx+1 + vx ) = 0 or, mx+2
1 vx = 0.
That is, 2 vx = 0. Since second dierence is zero, the rst dierence is constant and
hence vx is linear. Let vx = c1 + c2 x, where c1 , c2 are arbitrary constants.
Hence, in this case the general solution is
ux = (c1 + c2 x)mx1 .
Case III. If the roots m1 , m2 are complex, then m1 , m2 should be conjugate complex
and let them be ( + i) and ( i), where , are reals. Then the general solution
is
ux = c1 ( + i)x + c2 ( i)x .
 To simplify the above expression, substituting = r cos , = r sin , where r =
2 + 2 and tan = /.

58 Numerical Analysis
Therefore,
ux = c1 rx (cos + i sin )x + c2 rx (cos i sin )x
= rx {c1 (cos x + i sin x) + c2 (cos x i sin x)}
= rx {(c1 + c2 ) cos x + i(c1 c2 ) sin x}
= rx (A cos x + B sin x), where A = c1 + c2 and B = i(c1 c2 ).
Example 2.9.3 Solve ux+1 8ux = 0.
Solution. This equation is written as (E 8)ux = 0. Let ux = cm2 be a solution.
The A.E. is m 8 = 0 or, m = 8.
Then ux = c8x , where c is an arbitrary constant, is the general solution.
Example 2.9.4 Solve the dierence equation ux = ux1 +ux2 , x 2, u0 = 1, u1 =
1. Also, nd the approximate value of ux when x tends to a large number.
x
2
Solution. Let
ux = cm be a solution. The A.E. is m m 1 = 0
1 5
.
or, m =
2
Therefore, general solution is




1+ 5 x
1 5 x
u x = c1
+ c2
,
2
2

## where c1 , c2 are arbitrary constants.

Given that u0 = 1, u1 = 1, therefore,




1+ 5
1 5
+ c2
.
1 = c1 + c2 and 1 = c1
2
2
Solution of these equations is

5+1
1 5
and c2 = .
c1 =
2 5
2 5
Hence, the particular solution is





1 + 5 x+1
1
1 5 x+1
ux =
.

2
2
5
 1 5 x+1
0 and therefore,
When x then
2


1 1 + 5 x+1
.
ux 
2
5


59

## Solution of non-homogeneous dierence equation

The general form of non-homogeneous linear dierence equation with constant coecients is
(a0 E n + a1 E n1 + a2 E n2 + + an )ux = g(x), or, f (E)ux = g(x).
The solution of this equation is the combination of two solutions complementary
function (C.F.) and particular integral or particular solution (P.S.). The C.F. is
1
g(x).
the solution of the homogeneous equation f (E)ux = 0 and the P.I. is given by
f (E)
Then the general solution is
ux =C.F.+P.I.
The method to nd C.F. is discussed in previous section.
Rules to nd particular integral
Case I. g(x) = ax , f (a) = 0.
Since f (E) = a0 E n + a1 E n1 + + an ,
f (E)ax = a0 ax+n + a1 ax+n1 + + an ax = ax (a0 an + a1 an1 + an )
= ax f (a).
Thus P.I. =

1 x
1 x
a =
a , provided f (a) = 0.
f (E)
f (a)

## Case II. g(x) = ax (x). Then,

f (E)ax (x) = a0 E n ax (x) + a1 E n1 ax (x) + + an ax (x)
= ax [a0 an (x + n) + a1 an1 (x + n 1) + + an (x)]
= ax [(a0 an E n + a1 an1 E n1 + + an )(x)]
= ax f (aE)(x).
This gives P.I. =

1 x
1
a (x) = ax
(x), where f (aE) = 0.
f (E)
f (aE)

## Case III. g(x) = ax , f (a) = 0.

1
1 [by Case II]
In this case, P.I. = ax
f (aE)
Case IV. g(x) = xm (m is zero or positive integer)
1
1
xm =
xm .
Then, P.I. =
f (E)
f (1 + )
Now, this expression is evaluated by expanding [f (1 + )]1 as an innite series of
and applying dierent dierences on xm .

60 Numerical Analysis
Case V. g(x) = sin ax or cos ax.
1 iax
1
sin ax = Imaginary part of
e .
(i) P.I. =
f (E)
f (E)
(ii) P.I. =

1 iax
1
cos ax = Real part of
e .
f (E)
f (E)

## Example 2.9.5 Solve the equation ux+2 5ux+1 + 6ux = 5x + 2x .

Solution. The given equation is (E 2 5E + 6)ux = 5x + 2x .
Let ux = cmx be a solution of (E 2 5E +6)ux = 0. Therefore, A.E. is m2 5m+6 = 0
or, m = 2, 3.
Hence C.F. is c1 2x + c2 3x .
1
1
5x
5x =
5x = .
P.I. of 5x = 2
E 5E + 6
25 25 + 6
6
1
1
P.I. of 2x = 2
2x = 2x
1
E 5E + 6
(2E)2 5(2E) + 6
1
1
1 = 2x
1
= 2x 2
2
4E 10E + 6
4(1 + ) 10(1 + ) + 6
1
1
1 = 2x
(1 2)1 1
= 2x 2
4 2
2
1
1 = 2x1 x(1) = 2x1 .x
= 2x
2
5x
2x1 .x, where c1 , c2 are
Therefore, the general solution is ux = c1 2x + c2 3x +
6
arbitrary constants.
Example 2.9.6 Solve ux+2 4ux+1 + 3ux = 2x .x(3) .
Solution. The equation can be written as (E 2 4E + 3)ux = 2x x(3) . Let ux = cmx
be a solution. The A.E. is m2 4m + 3 = 0 or, m = 1, 3.
Therefore, C.F. is c1 1x + c2 3x = c1 + c2 3x .
1
1
2x x(3) = 2x
x(3)
2
4E + 3
(2E) 4(2E) + 3
1
x(3)
= 2x
2
4(1 + ) 8(1 + ) + 3
1
x(3) = 2x (1 42 )1 x(3)
= 2x 2
4 1
= 2x (1 + 42 + 164 + )x(3)

P.I. =

E2

## = 2x [x(3) + 4.3.2x(1) ] = 2x [x(3) + 24x(1) ].

Hence, the general solution is ux = c1 + c2 3x 2x [x(3) + 24x(1) ].

## Example 2.9.7 Solve ux ux1 + 2ux2 = x2 + 5x .

Solution. The given equation can be written as
(E 2 E + 2)ux2 = x2 + 5x .
x
2)ux2 = 0.
Let ux2 = cm be a solution of (E 2 E +
1

i
7
.
Then A.E. is m2 m + 2 = 0 or, m =
2

1
7
Here the roots are complex. Let = r cos and
= r sin .
2
2

Therefore, r = 2, tan = 7.

## The C.F. is ( 2)x [c1 cos x + c2 sin x] where tan = 7.

P.I. =
=
=
=
=
=
=

1
(x2 + 5x )
E+2
1
1
{x(2) + x(1) } +
5x
2
(1 + ) (1 + ) + 2
25 5 + 2
1
5x
(2)
(1)
{x
+
x
}
+
2 + + 2
22

1
2
1
+
5x
1+
{x(2) + x(1) } +
2
2
22



2
2
2
+
+
1
5x
1
+
{x(2) + x(1) } +
2
2
2
22

x
5
1 (2)
1
1
x + x(1) (2 + 2x(1) + 1) + (2) +
2
2
4
22
5x
5x
1 2
1 (2)
[x 1] +
= (x x 1) + .
2
22
2
22
E2

## Therefore, the general solution is

1
5x
ux2 = ( 2)x [c1 cos x + c2 sin x] + (x2 x 1) + ,
2
22

where
= 7 and c1 , c2 are arbitrary constants.
x
Example 2.9.8 Show that the solution of the equation ux+2 + ux = 2 sin
2

 x
x
+ x sin
.
is given by ux = a cos
2
2
Solution. Let ux = cmx be a solution of ux+2 + ux = 0.
Then A.E. is m2 + 1 = 0 or, m = i.
Therefore, C.F. is A(i)x + B(i)x .
Substituting 0 = r cos , 1 = r sin , where r = 1, = /2.

61

62 Numerical Analysis
Then C.F. reduces to
A{r(cos + i sin )}x + B{r(cos i sin )}x


x 
x 
x
x
+ i sin
i sin
= A cos
+ B cos
2
2
2
2
x
x
(B A)i sin
= (A + B) cos
2
2
x
x
= a cos cos
a sin sin
, where A + B = a cos , (B A)i = a sin
2 
2

x
+ .
= a cos
2
x
1
1
2 sin
= Imaginary part of 2
2eix/2
+1
2
E +1
1
1
[since (ei/2 )2 + 1 = 0]
= I.P. of 2eix/2 i/2 2
(e
E) + 1
1
1
= I.P. of 2eix/2 i 2
1
1 = I.P. of 2eix/2
e E +1
1 (1 + )2



ix/2 1
1 + 1
= I.P. of 2e
2
2
1
1
1 = I.P. of 2eix/2
x
= I.P. of 2eix/2
2
 2

x
x
= I.P. of (x) cos
+ i sin
2
2
x
= x sin
.
2

 x
x
+ x sin
.
Therefore, the general solution is ux = a cos
2
2
P.I. =

E2

## Example 2.9.9 Solve the following dierence equation

yn+2 4yn+1 + 4yn = n2 + 3n .
Solution. Let yn = cmn be a trial solution of (E 2 4E + 4)yn = 0.
Then A.E. is m2 4m + 4 = 0 or, m = 2, 2.
Therefore, C.F. is (c1 + c2 n)2n .
P.I. of 3n =

E2

1
1
3n =
3n = 3n .
4E + 4
(E 2)2

## To nd the P.I. of n2 , let yn = an2 + bn + c be the solution of yn+2 4yn+1 + 4yn = n2 .

Then {a(n + 2)2 + b(n + 2) + c} 4{a(n + 1)2 + b(n + 1) + c}
+4{an2 + bn + c} = n2 .

63

## That is, an2 + (b 4a)n + (c 2b) = n2 . Equating both sides we have

a = 1, b 4a = 0, c 2b = 0, i.e., a = 1, b = 4, c = 8.
Hence, the P.I. of n2 is n2 + 4n + 8.
Therefore, the general solution is
yn = (c1 + c2 n)2n + 3n + n2 + 4n + 8.
Example 2.9.10 Solve the dierence equation f (n) 4f (n) = 3,
n 2 and f (1) = 2.
Solution. The equation can be written using E operator as
(E 1)f (n) 4f (n) = 3

or,

(E 5)f (n) = 3.

## Let f (n) = cmn be a trial solution.

The A.E. is m 5 = 0. Therefore, C.F. is c5n .
To nd P.I. let, f (n) = a be the solution of the given equation.
Then (E 5)a = 3 or, a 5a = 3 or, a = 3/4.
Hence, the general solution is
3
f (n) = c5n .
4
But, given f (1) = 2. Therefore, 2 = 5c 3/4 or c = 11/20.
Hence the particular solution is
f (n) =

2.9.3

11 n1 3
.
5
4
4

Generating function

## Generating function is used to solve dierent kind of problems including combinatorial

problems. This function may also be used to solve dierence equation.
Def. 2.9.1 Let {a0 , a1 , a2 , . . .} be a sequence of real numbers. The power series
G(x) =

an xn

(2.65)

n=0

## is called the generating function for the sequence {a0 , a1 , a2 , . . .}.

In other words, an , the nth term of the sequence {an } is the coecient of xn in the
expansion of G(x). That is, if the generating function of a sequence is known, then one
can determine all the terms of the sequence.

64 Numerical Analysis

## Example 2.9.11 Use generating function to solve the dierence equation

un = 2un1 + 3, n 1, u0 = 2.
Solution. From the denition of the generating function,
G(x) =

un xn = u0 +

n=0

That is,

G(x) 2 =

un xn = 2 +

n=1

un xn .

n=1

un xn .

n=1

## Now, un = 2un1 + 3, n 1. Multiplying both sides by xn , we obtain

un xn = 2un1 xn + 3xn .
Taking summation for all n = 1, 2, . . ., we have

un xn = 2

n=1

That is,
G(x) 2 = 2x
= 2x


n=1

un1 xn + 3

n=1

un1 xn1 + 3

xn .

n=1

xn

n=1



un xn + 3
xn 1

n=0


= 2x G(x) + 3

n=0


1
1
1x


since

xn =

n=0

1 
1x

3
1.
1x
Therefore, the generating function for this dierence equation or for the sequence
{un } is

Thus,

(1 2x)G(x) =

3
1

(1 x)(1 2x) 1 2x
3
5

=
1 2x 1 x




n
n
=5
(2x) 3
x =
(5.2n 3)xn .

G(x) =

n=0

n=0

n=0

## The coecient of xn in the expansion of G(x) is 5.2n 3 and hence un = 5.2n 3 is

the required solution.

65

## Example 2.9.12 Using generating function solve the dierence equation an

5an1 + 6an2 = 0, n 2 with initial conditions a0 = 2 and a1 = 3.
Solution. Let G(x) be the generating function of the sequence {an }. Then



n
an x = 2 + 3x +
an xn .
G(x) = a0 + a1 x +
That is,

n=2

n=2

an x = G(x) 2 3x.
n

n=2

## Multiplying given equation by xn ,

an xn 5an1 xn + 6an2 xn = 0.
Taking summation for n = 2, 3, . . . ,


n=2

an xn 5

an1 xn + 6

n=2

or, G(x) 2 3x 5x

an2 xn = 0

n=2

n=2

or, G(x) 2 3x 5x




an xn a0

an2 xn2 = 0

n=2

+ 6x2 G(x) = 0

n=0

## or, G(x) 2 3x 5x[G(x) 2] + 6x2 G(x) = 0.

Therefore, G(x) =
ence equation.
Let

2 7x
. This is the generating function for the given dier1 5x + 6x2

B
(A + B) (3A + 2B)x
2 7x
A
+
=
.
=
2
1 5x + 6x
1 2x 1 3x
(1 2x)(1 3x)

## The unknown A and B are related by the equations

A + B = 2 and 3A + 2B = 7,
1
3

.
whose solution is A = 3, B = 1. Thus, G(x) =
1 2x 1 3x
Now,
G(x) = 3(1 2x)1 (1 3x)1




=3
(2x)n
(3x)n =
(3.2n 3n )xn .
n=0

n=0

n=0

## Hence an = coecient of xn in the expansion of G(x) = 3.2n 3n , which is the

required solution.

66 Numerical Analysis

2.10

Exercise

1. Dene the operators: forward dierence (), backward dierence (), shift (E),
central dierence () and average ().
2. Prove the following relations among the operators


2 2

2 2
, (ii) E 1/2 +
(i) 1 + 1 +
2
2
+
E 1
(iii)
, (iv)
+ ,
2
2
2
2


(v)
+ 1 + ( 2 /4), (vi) hD sinh1 (),
2
1
hD
(vii) hD 
log(1 +
 ) log(1 ) sinh (), (viii) E e ,

## (1 + )1/2 , (x) cosh(hD/2),

(ix) 1 +
2

2
2
2
2
(xi) + 1 + , (xii) E 1 +
+ 1+ ,
2
4
2
4
(xiii) E E 1/2 , (xiv) E 1 .
3. Show that
(i) i yk = i yk+i = i yk+i/2 , (ii) (yi2 ) = (yi + yi+1 )yi ,
1
yi
.
(iii) yi = yi = 2 yi , (iv)
=
yi
yi yi+1
4. Prove the following
n

(1)i n!f (x + (n i)h))
(i) n f (x) =
i!(n i)!
i=0
n
 (1)i n!f (x ih)
(ii) n f (x) =
i!(n i)!
(iii) 2n f (x) =

i=0
2n

i=0

.
i!(2n i)!

## 5. Prove the following

2 3
(i) hD
+

2
3
5
4
(ii) h2 D2 2 3 + 11
5
+
12
6
7
6
(iii) h4 D4 4 25 + 17
7
+ .
6
2
6. Prove that
(i) n (eax+b ) = (eah 1)n eax+b

## Calculus of Finite Diff. and Diff. Equs

67


2 x Eex
e 2 x
(ii) e =
E
e

f (x)
(iii) log f (x) = log 1 +
f (x)
 2

(iv)
x3 = 6x.
E


7. Prove that, if the spacing h is very small then the forward dierence operator is
almost equal to dierential operator, i.e., for small h, n f (x) hn Dn f (x).
8. Show that the operators , , E, and are commute with one another.
9. Express 3 yi and 4 y4 in terms of y.
10. Prove the following relations:
(i) ux = ux1 + ux2 + 2 ux3 + + n1 uxn + n uxn1
(ii) u1 + u2 + u3 + + un =n C1 u0 +n C2 u0 +n C3 2 u0 + + n1 u0
(iii) n yx = yn+x n C1 yx+n1 +n C2 yx+n2 + (1)n yx
(iv) u1 x + u2 x2 + u3 x3 +
x
x2
x3
=
u1 +
u
+
2 u1 + .
1
1x
(1 x)2
(1 x)3
11. Show that
(i) [f (x)g(x)] = f (x)g(x) + g(x + h)f (x)
(ii) n f (x) = n f (x + nh), where n is a positive integer
(iii) f (x) = f (x) f (x)
(iv) f (x) + f (x) = (/)f (x) (/)f (x).
12. The nth dierence n be dened as n f (x) = n1 f (x+h)n1 f (x), (n 1).
If f (x) is a polynomial of degree n show that f (x) is a polynomial of degree n1.
Hence deduce that the nth dierence of f (x) is a constant.
13. If r (x) = (x x0 )(x x1 ) (x xr ) where xr = x0 + rh, r = 0, 1, 2, . . . , n,
calculate k r (x).
14. If fi is the value of f (x) at xi where xi = x0 + ih, i = 1, 2, . . . prove that
fi = E i f0 =

i  

i
j=0

j f0 .

## 15. For equally spaced points x0 , x1 , . . . , xn , where xk = x0 + kh, (h > 0, k =

0, 1, . . . , n) express k y0 in terms of the ordinates.

68 Numerical Analysis
16. Taking h = 1, compute the second, third and fourth dierences of f (x) = 3x4
2x2 + 5x 1.
17. Construct the forward dierence table for the following tabulated values of f (x)
and hence nd the values of 2 f (3), 3 f (2), 4 f (0).
x
f (x)

:
:

0
4

1
7

2
10

3
20

4
45

5
57

6
70

18. Use nite dierence method to nd a polynomial which takes the following values:
x
f (x)

:
:

2
12

1
6

0
0

1
6

2
10

## 19. Compute the missing term in the following table.

x
f (x)

:
:

0
12

2
6

4
0

6
?

8
25

20. Use nite dierence method to nd the value of f (2.2) from the following data.
x
f (x)

:
:

1
3

2
24

3
99

4
288

5
675

## 21. Find the functions, whose rst dierences are

(i) 3x2 + 9x + 2, (ii) x4 3x3 + x2 11x + 20.
22. Use iteration method to solve the following dierence equations
(i) an = an1 + 4, for all n 2 with a1 = 2.
(ii) un = un1 + (n 1), n 2 and a1 = 0.
(iii) xn = 5xn1 + 3 for n 2 and x1 = 2.
23. Find the rst ve terms of the sequence dened by the following recurrence relations:
(i) xn = x2n1 for n 2 and x1 = 1.
(ii) xn = nxn1 + n2 xn2 for n 2 and x0 = 1, x1 = 1.
(iii) Let x1 = 1 and for n 2, xn = x1 xn1 + x2 xn2 + + xn1 x1 .
(The numbers of this sequence are called Catalan numbers).
24. Mr. Das deposits Rs. 1,000 in a bank account yielding 5% compound interest
yearly.
(i) Find a recurrence relation for the amount in the account after n years.
(ii) Find an explicit formula for the amount in the account after n years.
(iii) How much will be in the account after 10 years ?

69

## 25. Solve the following dierence equations.

(i) un 5un1 + 6un2 = n2 + 7n + 3n
(ii) ux+2 7ux1 + 10ux = 12e3x + 4x
(iii) un+2 4un+1 + 4un = n for n 1 and u1 = 1, u2 = 4
(iv) un 5un1 + 6un2 = n2 + 5n + 2n
(v) 6un+2 7un+1 20un = 3n2 2n + 8
(vi) f (n + 2) 8f (n + 1) + 25f (n) = 2n2 + n + 1
(vii) ux ux1 + 2ux2 = x + 2x
(viii) yn+2 4yn+1 + 4yn = n + 3n
(ix) un 5un1 + 6un2 = n2
(x) Sn+1 = Sn + Sn1 , n 3 and S1 = 1, S2 = 1. Find S8 .
26. Assuming un = an + b, show that the particular solution of
1
un 5un1 + 6un2 = n is (2n + 7).
4
27. Show that the general solution of ux ux1 ux2 = x2

1
is x [A(1 + 5)x + B(1 5)x ] (x2 + 6x + 13).
2
2
28. Show that
 the solution of ux+2
 + a 2ux = cos ax is
x
x
a cos ax + cos a(2 x)
.
+ B sin
+
ux = ax A cos
2
2
1 + 2a2 cos 2a + a4





11 + a n+1
1
11 a n+1

## u1 = 11 then show that un is given by un =

a
2
2

where a = 3 17.
30. The seeds of a certain plant when one year old produce eighteen fold. A seed is
planted and every seed subsequently produced is planted as soon as it is produced.
Prove that the number of grain at the end of nth year is
1
un =
a

11 + a
2

n+1

11 a
2

n+1

where a = 3 17.
31. The rst term of a sequence {un } is 1, the second is 4 and every other term is the
arithmetic mean of the two preceding terms. Find un and show that un tends to
a denite limit as n .
32. The rst term of a sequence is 1, the second is 2 and every term is the sum of the
two proceeding terms. Find the nth term.

70 Numerical Analysis
33. If ur satises the dierence equation ur 4ur1 + ur2 = 0, 2 r n, where

A sinh(n r)
.
un = 0 and u0 = A, show that, if = log(2 + 3) then ur =
sinh n
34. Show that the general solution of the national income equation
1
1
yn+2 yn+1 yn = nh + A where h, A are constants, is given by
2n
4
yn = c1 m1 + c2 mn2 + 4hn + 4(A 6h) where c1 , c2 are arbitrary constants and the
values of m1 and m2 you are to actually nd out. Also show that yn /n tends to
nite limit as n .
35. Use generating functions to solve the following dierence equations.
(i) xn = 3xn1 , n 1 and x0 = 2.
(ii) xn = 5xn1 + 2n , n 1 and x0 = 2.
(iii) xn = xn1 + n for n 1 and x0 = 1.
(iv) un un1 un2 = 0 for n 2 and u0 = 0, u1 = 1.
(v) an + an2 = 2an1 , n 2 and a0 = 0, a1 = 1.

Chapter 3

Interpolation
Sometimes we have to compute the value of the dependent variable for a given independent variable, but the explicit relation between them is not known. For example,
the Indian population are known to us for the years 1951, 1961, 1971, 1981, 1991 and
2001. There is no exact mathematical expression available which will give the population for any given year. So one can not determine the population of India in the year
2000 analytically. But, using interpolation one can determine the population (obviously
approximate) for any year.
The general interpolation problem is stated below:
Let y = f (x) be a function whose analytic expression is not known, but a table of
values of y is known only at a set of values x0 , x1 , x2 , . . ., xn of x. There is no other
information available about the function f (x). That is,
f (xi ) = yi , i = 0, 1, . . . , n.
(3.1)
The problem of interpolation is to nd the value of y(= f (x)) for an argument, say,
x . The value of y at x is not available in the table.
A large number of dierent techniques are used to determine the value of y at x = x .
But one common step is nd an approximate function, say, (x), corresponding to
the given function f (x) depending on the tabulated value. The approximate function
should be simple and easy to handle. The function (x) may be a polynomial, exponential, geometric function, Taylors series, Fourier series, etc. When the function (x) is
a polynomial, then the corresponding interpolation is called polynomial interpolation.
The polynomial interpolation is widely used interpolation technique, because, polynomials are continuous and can be dierentiated and integrated term by term within any
range.
A polynomial (x) is called interpolating polynomial if yi = f (xi ) = (xi ), i =
dk f 
dk 
0, 1, 2, . . . , n and
=
for some nite k, and x is one of the values of x0 ,
dxk x
dxk x
71

72 Numerical Analysis
y

y = f (x) +

y = (x)
:

y=f (x)

y = f (x)

-x

x0 x1

x2

xn

## Figure 3.1: Interpolation of a function.

x1 , . . ., xn .
The following theorem justies the approximation of an unknown function f (x) to a
polynomial (x).
Theorem 3.1 If the function f (x) is continuous in [a, b], then for any pre-assigned
positive number > 0, there exists a polynomial (x) such that
|f (x) (x)| < for all x (a, b).
This theorem ensures that the interpolating polynomial (x) is bounded by y =
f (x) and y = f (x) + , for a given . This is shown in Figure 3.1.
Depending on the tabulated points, several interpolation methods are developed.
Among them nite-dierence interpolating formulae, Lagranges interpolation are widely
used polynomial interpolation methods.
For the sake of convenience, a polynomial of degree n, means a polynomial of degree
not higher than n.

3.1

## Lagranges Interpolation Polynomial

Let y = f (x) be a real valued function dened on an interval [a, b]. Let x0 , x1 , . . . , xn be
n + 1 distinct points in the interval [a, b] and y0 , y1 , . . . , yn be the corresponding values
of y at these points, i.e., yi = f (xi ), i = 0, 1, . . . , n, are given.

Interpolation

73

## Now, we construct an algebraic polynomial (x) of degree less than or equal to n

which attains the assigned values at the points xi , that is,
(xi ) = yi , i = 0, 1, . . . , n.

(3.2)

The polynomial (x) is called the interpolation polynomial and the points xi ,
i = 0, 1, . . . , n are called interpolation points.
Let the polynomial (x) be of the form
(x) =

n


Li (x) yi ,

(3.3)

i=0

where each Li (x) is polynomial in x, of degree less than or equal to n, called the
Lagrangian function.
The polynomial (x) satises the equation (3.2) if

0, for i = j
Li (xj ) =
1, for i = j.
That is, the polynomial Li (x) vanishes only at the points x0 , x1 , . . . , xi1 , xi+1 , . . . , xn .
So it should be of the form
Li (x) = ai (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn ),
where ai , a constant whose value is determined by using the relation
Li (xi ) = 1.
Then ai (xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn ) = 1.
or, ai = 1/{(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )}.
Therefore,
Li (x) =

## (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn )

.
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )

(x) =

n


Li (x) yi ,

i=0

## where Li (x) is given in (3.4).

The polynomial Li (x) can be written as
Li (x) =


n 

x xj
.
xi xj
j=0
j=i

(3.4)

74 Numerical Analysis
In this notation, the polynomial (x) is


n 
n 

x xj
yi .
(x) =
xi xj
j=0
i=0

j=i

## The function Li (x) can also be expressed in another form as follows.

Let
w(x) = (x x0 )(x x1 ) (x xn )

(3.5)

## be a polynomial of degree n + 1 and vanishes at x = x0 , x1 , . . . , xn .

Now, the derivative of w(x) with respect to x is given by
w (x) = (x x1 )(x x2 ) (x xn ) + (x x0 )(x x2 ) (x xn )
+ + (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn )
+ + (x x0 )(x x1 ) (x xn1 ).
Therefore, w (xi ) = (xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn ), which is
the denominator of Li (x).
Using w(x), Li (x) becomes
Li (x) =

w(x)
.
(x xi )w (xi )

(x) =

n

i=0

w(x)
yi .
(x xi )w (xi )

(3.6)

## Example 3.1.1 Obtain Lagranges interpolating polynomial for f (x) and nd an

approximate value of the function f (x) at x = 0, given that f (2) = 5, f (1) = 1
and f (1) = 1.
Solution. Here x0 = 2, x1 = 1, x2 = 1 and f (x0 ) = 5, f (x1 ) = 1, f (x2 ) = 1.
2

Then f (x) 
Li (x)f (xi ).
i=0

Now,

(x + 1)(x 1)
x2 1
(x x1 )(x x2 )
=
=
.
(x0 x1 )(x0 x2 )
(2 + 1)(2 1)
3
(x + 2)(x 1)
x2 + x 2
(x x0 )(x x2 )
L1 (x) =
=
=
.
(x1 x0 )(x1 x2 )
(1 + 2)(1 1)
2
(x + 2)(x + 1)
x2 + 3x + 2
(x x0 )(x x1 )
=
=
.
L2 (x) =
(x2 x0 )(x2 x1 )
(1 + 2)(1 + 1)
6
L0 (x) =

Interpolation

75

Therefore,
x2 + x 2
x2 + 3x + 2
x2 1
(5) +
(1) +
1
3
2
6
= 1 + x x2 .

f (x) 

Thus, f (0) = 1.
The Lagrangian coecients can be computed from the following scheme. The dierences are computed, row-wise, as shown below:
x x0
x1 x0
x2 x0

xn x0

x0 x1
x x1
x2 x1

xn x1

x0 x2
x1 x2
x x2

x x2

x0 xn
x1 xn
x2 xn

x xn

From this table, it is observed that the product of diagonal elements is w(x). The
product of the elements of rst row is (x x0 )w (x0 ), product of elements of second row
is (x x1 )w (x1 ) and so on. Then the Lagrangian coecient can be computed using
the formula
w(x)
.
Li (x) =
(x xi )w (xi )
Linear Lagrangian Interpolation
Let x0 and x1 be two points and y0 and y1 be the corresponding values of y. In this
case,
(x) = L0 (x)y0 + L1 (x)y1
x x1
x x0
x x0
=
y0 +
y1 = y0 +
(y1 y0 ).
x0 x1
x1 x0
x1 x0

(3.7)

3.1.1

## For the equally spaced points, xi = x0 + ih, i = 0, 1, 2 . . . , n, where h is the spacing.

Now, a new variable s is introduced, dened by x = x0 + sh.
Then x xi = (s i)h and xi xj = (i j)h.
Therefore,
w(x) = (x x0 )(x x1 ) (x xn ) = sh(s 1)h(s 2)h (s n)h
= hn+1 s(s 1)(s 2) (s n).

76 Numerical Analysis
Also,
w (xi ) = (xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 )(xi xi+2 ) (xi xn )
= (ih)(i 1)h (i i 1)h(i i + 1)h(i i + 2)h (i n)h
= hn i(i 1) 1 (1)(2) ({(n i)}
= hn i!(1)ni (n i)!.
Using these values, the relation (3.5) becomes
(x) =
=

n

hn+1 s(s 1)(s 2) (s n)
i=0
n


## (1)ni hn i!(n i)!(s i)h

(1)ni

i=0

yi

s(s 1)(s 2) (s n)
yi ,
i!(n i)!(s i)

(3.8)

where x = x0 + sh.
For given tabulated values, the Lagranges interpolation polynomial exists and unique.
These are proved in the following theorem.
Theorem 3.2 The Lagranges interpolation polynomial exists and unique.
Proof. The Lagranges interpolation formula satised the condition
yi = (xi ), i = 0, 1, . . . , n.

(3.9)

For n = 1,
(x) =

x x0
x x1
y0 +
y1 .
x0 x1
x1 x0

(3.10)

For n = 2,
(x) =

(x x0 )(x x2 )
(x x1 )(x x2 )
y0 +
y1
(x0 x1 )(x0 x2 )
(x1 x0 )(x1 x2 )
(x x0 )(x x1 )
y2 .
+
(x2 x0 )(x2 x1 )

(3.11)

(x) =

n

i=0

Li (x)yi ,

(3.12)

Interpolation

77

where
Li (x) =

## (x x0 ) (x xi1 )(x xi+1 ) (x xn )

,
(xi x0 ) (xi xi1 )(xi xi+1 ) (xi xn )
i = 0, 1, . . . , n.

(3.13)

Expression (3.10) is a linear function, i.e., a polynomial of degree one and also,
(x0 ) = y0 and (x1 ) = y1 .
Also, expression (3.11) is a second degree polynomial and (x0 ) = y0 , (x1 ) =
y1 , (x2 ) = y2 , i.e., satisfy (3.9). Thus, the condition (3.13) for n = 1, 2 is fullled.
The functions (3.13) expressed in the form of a fraction whose numerator is a polynomial of degree n and whose denominator is a non-zero number. Also, Li (xi ) = 1 and
Li (xj ) = 0 for j = i, j = 0, 1, . . . , n. That is, (xi ) = yi . Thus, the conditions of (3.9)
are satised. Hence, the Lagranges polynomial exists.
Uniqueness of the polynomial
Let (x) be a polynomial of degree n, where
(xi ) = yi , i = 0, 1, . . . , n.

(3.14)

## Also, let (x) be another polynomials of degree n satisfying the conditions

(xi ) = yi , i = 0, 1, . . . , n.

(3.15)

## Then from (3.14) and (3.15),

(xi ) (xi ) = 0, i = 0, 1, . . . , n.

(3.16)

If (x) (x) = 0, then this dierence is a polynomial of degree at most n and it has
at most n zeros, which contradicts (3.16), whose number of zeros is n + 1. Consequently,
(x) = (x). Thus (x) is unique.

3.2

## 1. The Lagrangian functions depend only on xi s and independent of yi s or f (xi )s.

2. The form of Lagrangian functions remain unchanged (invariant) under linear transformation.
Proof. Let x = az + b, where a, b are arbitrary constants.
Then xj = azj + b and x xj = a(z zj ). Also, xi xj = a(zi zj )
(i = j).

78 Numerical Analysis
Therefore,
w(x) = (x x0 )(x x1 ) (x xn )
= an+1 (z z0 )(z z1 ) (z zn ).
w (xi ) = (xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )
= an (zi z0 )(zi z1 ) (zi zi1 )(zi zi+1 ) (zi zn ).
Thus,
Li (x) =

w(x)
(x xi )w (xi )

an+1 (z z0 )(z z1 ) (z zn )
a(z zi )an (zi z0 )(zi z1 ) (zi zi1 )(zi zi+1 ) (zi zn )
w(z)
= Li (z).
=
(z zi )w (zi )
=

## Thus Li (x)s are invariant.

3. Sum of Lagrangian functions is 1, i.e.,

n


Li (x) = 1.

i=0

n

i=0

Li (x) =

n

i=0

w(x)
(x xi )w (xi )

(3.17)

## where w(x) = (x x0 )(x x1 ) (x xn ).

Let
A0
A1
Ai
An
1
=
+
+ +
+ +
(3.18)
w(x)
x x0 x x1
x xi
x xn
i.e., 1 = A0 (x x1 )(x x2 ) (x xn ) + A1 (x x0 )(x x2 ) (x xn )
+ + Ai (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn )
+ + An (x x0 )(x x1 ) (x xn1 ).
When x = x0 is substituted in (3.19) then the value of A0 is given by
1 = A0 (x0 x1 )(x0 x2 ) (x0 xn )
That is,
1
1
A0 =
= 
.
(x0 x1 )(x0 x2 ) (x0 xn )
w (x0 )
Similarly, x = x1 gives

(3.19)

Interpolation

A1 =

1
w (x1 )

79

1
1
and An = 
.
w (xi )
w (xn )
Using these results, equation (3.18) becomes
Also, Ai =

1
1
1
=
+
+

w(x)
(x x0 )w (x0 ) (x x1 )w (x1 )
1
1
+ +
+

(x xi )w (xi )
(x xn )w (xn )
n

w(x)
.
i.e., 1 =
(x xi )w (xi )
i=0

n

i=0

3.3

Li (x) =

n

i=0

w(x)
= 1.
(x xi )w (xi )

(3.20)

## It is obvious that if f (x) is approximated by a polynomial (x), then there should be

some error at the non-tabular points. The following theorem gives the amount of error
in interpolating polynomial.
Theorem 3.3 Let I be an interval contains all interpolating points x0 , x1 , . . . , xn . If
f (x) is continuous and have continuous derivatives of order n + 1 for all x in I then
the error at any point x is given by
En (x) = (x x0 )(x x1 ) (x xn )

f (n+1) ()
,
(n + 1)!

(3.21)

where I.
Proof. Let the error En (x) = f (x) (x), where (x) is a polynomial of degree less
than or equal to n, which approximates the function f (x).
Now, En (xi ) = f (xi ) (xi ) = 0 for i = 0, 1, . . . , n.
By virtue of the above result, it is assumed that En (x) = w(x)k, where w(x) =
(x x0 )(x x1 ) . . . (x xn ).
The error at any point, say, x = t, other than x0 , x1 , . . . , xn is En (t) = w(t)k
or, f (t) (t) = kw(t).

(3.22)

80 Numerical Analysis
Let us construct an auxiliary function
F (x) = f (x) (x) kw(x).

(3.23)

## The function vanishes at x = x0 , x1 , . . . , xn because f (xi ) = (xi ) and w(xi ) = 0.

Also, F (t) = 0, by (3.22).
Hence, F (x) = 0 has n + 2 roots in I. By Rolls theorem, F  (x) = 0 has n + 1 roots
in I. F  (x) = 0 has n roots in I and nally, F (n+1) (x) = 0 must have at least one root
in I. Let be one such root. Then F (n+1) () = 0. That is, f (n+1) () 0 + k(n + 1)! = 0
[(x) is a polynomial of degree n so (n+1) (x) = 0 and w(x) is a polynomial of degree
n + 1, so w(n+1) (x) = (n + 1)!]. Thus
k=

f (n+1) ()
.
(n + 1)!

## Therefore, the error at x = t is

En (t) = kw(t) [by (3.22)]
= (t x0 )(t x1 ) (t xn )

f (n+1) ()
.
(n + 1)!

## Hence, the error at any point x is

En (x) = (x x0 )(x x1 ) (x xn )

f (n+1) ()
.
(n + 1)!

Note 3.3.1 The above expression gives the error at any point x.
But practically it has little utility, because, in many cases f (n+1) () cannot be determined.
If Mn+1 be the upper bound of f (n+1) () in I, i.e., if |f (n+1) ()| Mn+1 in I then
the upper bound of En (x) is
|En (x)|

Mn+1
|w(x)|.
(n + 1)!

(3.24)

## Note 3.3.2 (Error for equispaced points)

Let xi = x0 + ih, i = 0, 1, . . . , n and x = x0 + sh then x xi = (s i)h.
Then the error is
En (x) = s(s 1)(s 2) (s n)hn+1

f (n+1) ()
.
(n + 1)!

(3.25)

Interpolation

81

Note 3.3.3 (Error bounds for equally spaced points, particular cases)
Assume that, f (x) is dened on [a, b] that contains the equally spaced points. Suppose,
f (x) and the derivatives up to n + 1 order are continuous and bounded on the intervals
[x0 , x1 ], [x0 , x2 ] and [x0 , x3 ] respectively. That is, |f (n+1)() | Mn+1 for x0 xn ,
for n = 1, 2, 3. Then
h2 M2
,
8
h3 M3
(ii) |E2 (x)| ,
9 3
h4 M4
(iii) |E3 (x)|
,
24
(i) |E1 (x)|

x0 x x1

(3.26)

x0 x x2

(3.27)

x0 x x3 .

(3.28)

## Proof. (i) From (3.25),

|E1 (x)| = |s(s 1)|h2

|f (2) ()|
.
2!

Let g1 (s) = s(s 1). g1 (s) = 2s 1. Then s = 1/2, which is the solution of g1 (s) = 0.
The extreme value of g1 (s) is 1/4.
Therefore,
h2 M2
1 M2
=
.
|E1 (x)| h2
4
2!
8
|f (3) ()|
.
3!
1
Let g2 (s) = s(s 1)(s 2). Then g2 (s) = 3s2 6s + 2. At g2 (s) = 0, s = 1 .
3
Again g2 (s) = 6(s 1) < 0 at s = 1 13 .
Therefore, the maximum value of g2 (s) is
(ii) |E2 (x)| = |s(s 1)(s 2)|h3


2
1
1 
1 
1 = .

1
3
3 3
3
3

Thus,
h3 M3
2
M3
|E2 (x)| h3
= .
6
3 3
9 3
|f (4 ()|
.
4!

Let g3 (s) = s(s 1)(s 2)(s 3). Then g3 (s) = 4s3 18s2 + 22s 6.
i.e., 2s3 9s2 + 11s 3 = 0.
At extrema, g3 (s) = 0,
3 3 5
.
This gives s = ,
2
2
(iii) |E3 (x)| = |s(s 1)(s 2)(s 3)|h4

82 Numerical Analysis

g3 (s)

g3 (3/2)

g3

 3 5 

> 0.
18s + 11. Then
< 0 and
2

3
3 5
9
But, |g3 (s)| = 1 at s =
and |g3 (s)| =
at x = .
2
16
2
Therefore the maximum value of |g3 (s)| is 1.
Hence,
h4 M4
M4
=
.
|E3 (x)| 1.h4
24
24
=

6s2

## Comparison of accuracy and O(hn+1 )

The equations (3.26), (3.27) and (3.28) give the bounds of errors for linear, quadratic
and cubic interpolation polynomials. In each of these cases the error bound |En (x)|
depends on h in two ways.
Case I. hn+1 is present explicitly in |En (x)| and En (x) is proportional to hn+1 .
Case II. The value of Mn+1 generally depends on the choice of h and tend to |f (n+1) (x0 )|
as h goes to zero.
Therefore, as h tends to zero |En (x)| converges to zero with the same rate that hn+1
converges to zero. Thus, one can say that |En (x)| = O(hn+1 ). In particular,
|E1 (x)| = O(h2 ), |E2 (x)| = O(h3 ), |E3 (x)| = O(h4 ) and so on.
As a consequence, if the derivatives of f (x) are uniformly bounded on the interval
and |h| < 1, then there is a scope to choose n suciently large to make hn+1 very small,
and the higher degree polynomial will have less error.
Example 3.3.1 Consider f (x) = cos x over [0, 1.5]. Determine the error bounds
for linear, quadratic and cubic Lagranges polynomials.
Solution. |f  (x)| = | sin x|, |f  (x)| = | cos x|, |f  (x)| = | sin x|,
|f iv (x)| = | cos x|.
|f  (x)| | cos 0| = 1.0, so that M2 = 1.0,
|f  (x)| | sin 1.5| = 0.997495, so that M3 = 0.997495,
|f iv (x)| | cos 0| = 1.0, so that M4 = 1.0.
For linear polynomial the spacing h of the points is 1.5 0 = 1.5 and its error bound
is
(1.5)2 1.0
h2 M2

= 0.28125.
|E1 (x)|
8
8
For quadratic polynomial the spacing of the points is h = (1.5 0)/2 = 0.75 and its
error bound is
|E2 (x)|

(0.75)3 0.997495
h3 M5

= 0.0269955.
9 3
9 3

Interpolation

83

The spacing for cubic polynomial is h = (1.5 0)/3 = 0.5 and thus the error bound
is
(0.5)4 1.0
h4 M4

= 0.0026042.
|E3 (x)|
24
24

Example 3.3.2 A function f (x) dened on the interval (0, 1) is such that f (0) =
0, f (1/2) = 1, f (1) = 0. Find the quadratic polynomial p(x) which agrees with f (x)
for x = 0, 1/2, 1.
 d3 f 
1


for 0 x 1.
If  3  1 for 0 x 1, show that |f (x) p(x)|
dx
12
Solution. Given x0 = 0, x1 = 1/2, x2 = 1 and f (0) = 0, f (1/2) = 1, f (1) = 0.
From Lagranges interpolating formula, the required quadratic polynomial is
(x x0 )(x x2 )
(x x1 )(x x2 )
f (x0 ) +
f (x1 )
(x0 x1 )(x0 x2 )
(x1 x0 )(x1 x2 )
(x x0 )(x x1 )
f (x2 )
+
(x2 x0 )(x2 x1 )
(x 0)(x 1)
(x 0)(x 1/2)
(x 1/2)(x 1)
0+
(1) +
0
=
(0 1/2)(0 1)
(1/2 0)(1/2 1)
(1 0)(1 1/2)
= 4x(x 1).

p(x) =

## The error E(x) = f (x) p(x) is given by



f ()
E(x) = (x x0 )(x x1 )(x x2 )
3!
 f  () 


or, |E(x)| = |x x0 ||x x1 ||x x2 |

3!

 d3 f 
1


as  3  1 in 0 x 1 .
|x 0||x 1/2||x 1|1.
3!
dx
Now, |x 0| 1, |x 1/2| 1/2 and |x 1| 1 in 0 x 1.
1
1 1
Hence, |E(x)| 1. .1. = .
2 6
12
1
That is, |f (x) p(x)| .
12
Example 3.3.3 Determine the step size h (and number of points n) to be used in
the tabulation of f (x) = cos x in the interval [1, 2] so that the quadratic interpolation
will be correct to six decimal places.

84 Numerical Analysis
Solution. The upper bound of error in quadratic polynomial is
|E2 (x)|
f (x) = cos x,

h3 M3
,
9 3

f  (x) = sin x,

## M3 = max f  (x).

1x2

f  (x) = cos x,

## max |f  (x)| = max | sin x| = 1.

1x2

h3
Hence 1 5 106 ,
9 3
This gives h 0.0427 and n =

1x2

i.e., h3 45 3 106 .
21
= 23.42  24.
h

In Lagrangian interpolation, there is no restriction on the spacing and order of the
tabulating points x0 , x1 , . . . , xn . Also, the value of y (the dependent variable) can be
calculated at any point x within the minimum and maximum values of x0 , x1 , . . . , xn .
But its main disadvantage is, if the number of interpolating points decreases or increases then fresh calculation is required, the previous computations are of little help.
This disadvantage is not in Newtons dierence interpolation formulae, which are discussed in Section 3.5.
Example 3.3.4 Obtain a quadratic polynomial approximation to f (x) = ex using
Lagranges interpolation method, taking three points x = 0, 1/2, 1.
Solution. Here x0 = 0, x1 = 1/2, x2 = 1 and f (x0 ) = 1, f (x1 ) = e1/2 ,
f (x2 ) = e1 .
(x x0 )(x x2 )
(x x1 )(x x2 )
f (x0 ) +
f (x1 )
(x0 x1 )(x0 x2 )
(x1 x0 )(x1 x2 )
(x x0 )(x x1 )
f (x2 )
+
(x2 x0 )(x2 x1 )
(x 0)(x 1)
(x 0)(x 1/2)
(x 1/2)(x 1)
1+
e1/2 +
e1
=
(0 1/2)(0 1)
(1/2 0)(1/2 1)
(1 0)(1 1/2)

(x) =

## = (2x 1)(x 1) 4e1/2 x(x 1) + e1 x(2x 1)

= 2x2 (1 2e1/2 + e1 ) x(3 4e1/2 + e1 ) + 1
= 0.309636243x2 0.941756802x + 1.0.
The functions f (x) and (x) are shown in Figure 3.2.

Interpolation

85

y
y=f (x)

y=(x)

- x

Figure 3.2: The graph of the function f (x) = ex and the polynomial
(x) = 0.309636243x2 0.941756802x + 1.

## Algorithm 3.1 (Lagranges interpolation). This algorithm determines the value

of y at x = x , say, from a given table of points (xi , yi ), i = 0, 1, 2, . . . , n, using
Lagranges interpolation method.
Algorithm Lagrange Interpolation
Step 1: Read x, n // n represents the number of points minus one//
// x is the interpolating point//
Step 2: for i = 0 to n do
endfor;
Step 3: Set sum = 0;
Step 4: for i = 0 to n do
Step 4.1: Set prod = 1;
Step 4.2: for j = 0 to n do
x xj
;
if (i = j) then prod = prod
xi xj
Step 4.3: Compute sum = sum + yi prod;
endfor;
Step 5: Print x, sum;
end Lagrange Interpolation
Program 3.1
.
/* Program Lagrange Interpolation
This program implements Lagranges interpolation
formula for one dimension; xg is the interpolating points */

86 Numerical Analysis

#include <stdio.h>
#include <math.h>
void main()
{
int n, i, j; float xg, x, y, sum=0, prod=1;
printf("Enter the value of n and the data
in the form x[i],y[i] ");
scanf("%d",&n);
for(i=0;i<=n;i++) scanf("%f %f",&x[i],&y[i]);
printf("\nEnter the interpolating point x ");
scanf("%f",&xg);
for(i=0;i<=n;i++)
{
prod=1;
for(j=0;j<=n;j++)
{
if(i!=j) prod*=(xg-x[j])/(x[i]-x[j]);
}
sum+=y[i]*prod;
}
printf("\nThe given data is ");
for(i=0;i<=n;i++) printf("\n(%6.4f,%6.4f)",x[i],y[i]);
printf("\nThe value of y at x= %5.2f is %8.5f ", xg, sum);
} /* main */
A sample of input/output:
Enter the value of n and the data in the form x[i],y[i] 4
1
5
1.5 8.2
2
9.2
3.2 11
4.5 16
Enter the interpolating point x 1.75
The given data is
(1.0000,5.0000)
(1.5000,8.2000)
(2.0000,9.2000)
(3.2000,11.0000)
(4.5000,16.0000)
The value of y at x= 1.75 is 8.85925

Interpolation

3.4

87

Finite Dierences

Dierent types of nite dierences are introduced in Chapter 2. Some of them are
recapitulated here.
Let a function y = f (x) be known as (xi , yi ) at (n+1) points xi , i = 0, 1, . . . , n, where
xi s are equally spaced, i.e., xi = x0 + ih, h is the spacing between any two successive
points xi s. That is, yi = f (xi ), i = 0, 1, . . . , n.
3.4.1

Forward dierences

## The rst forward dierence of f (x) is dened as

f (x) = f (x + h) f (x),
is called forward dierence operator.
Then f (x0 ) = f (x0 + h) f (x0 ) = f (x1 ) f (x0 ).
That is, y0 = y1 y0 using yi = f (xi ).
Similarly, y1 = y2 y1 , y2 = y3 y2 , . . ., yn1 = yn yn1 .
The second order dierences are
2 y0 = (y0 ) = (y1 y0 )
= y1 y0 = (y2 y1 ) (y1 y0 ) = y2 2y1 + y0 .
Similarly, 2 y1 = y3 2y2 + y1 , etc.
The third order dierences
3 y0 = (2 y0 ) = (y2 2y1 + y0 ) = y3 3y2 + 3y1 y0 ,
3 y1 = y4 3y3 + 3y2 y1 , etc.
In general,
k y0 = yk k C1 yk1 + k C2 yk2 (1)k y0

(3.29)

## yi = yk+i C1 yk+i1 + C2 yk+i2 (1) yi .

k

(3.30)

It is observed that dierence of any order can easily be expressed in terms of the
ordinates yi s with binomial coecients.
All orders forward dierences can be written in a tabular form shown in Table 3.1.
This dierence table is called forward dierence table or diagonal dierence
table.

88 Numerical Analysis

## Table 3.1: Forward dierence table.

x
x0

y
y0

2 y

3 y

4 y

y0
x1

2 y0

y1

3 y0

y1
x2

2 y1

y2
y2

x3

y3

x4

y4

4 y0
3 y

2 y2
y3

3.4.2

Backward dierences

## The rst order backward dierence of f (x) is dened as

f (x) = f (x) f (x h),
where is the backward dierence operator.
Thus, f (x1 ) = f (x1 ) f (x0 ), or, y1 = y1 y0 .
Similarly, y2 = y2 y1 , y3 = y3 y2 , . . ., yn = yn yn1 .
The second order dierences are
2 y2 = (y2 ) = (y2 y1 ) = y2 2y1 + y0 ,
2 y3 = y3 2y2 + y1 , etc.
The third order dierences are
3 y3 = y3 3y2 + 3y1 y0 ,
3 y4 = y4 3y3 + 3y2 y1 , etc.
In general,

## k yi = yi k C1 yi1 + k C2 yi2 (1)k yik .

(3.31)

Table 3.2 shows how the backward dierences of all orders can be formed.
The backward dierence table is sometimes called horizontal dierence table.
3.4.3

## Error propagation in a dierence table

If there is an error in any entry among the tabulated values of a function, then this
error propagates to other entries of higher order dierences. To illustrate the behaviour
of propagation of error, we assume that an error is present in the number, say, y3 .

Interpolation

89

x
x0
x1
x2
x3
x4

y
y0
y1
y2
y3
y4

2 y

3 y

4 y

y1
y2
y3
y4

2 y2
2 y3
2 y4

3 y3
3 y4

4 y4

## Table 3.3: Error propagation in a nite dierence table.

x
x0

y
y0

x1

y1

2 y

3 y

4 y

5 y

y0
2 y0
3 y0 +

y1
x2

2 y

y2
y2 +

x3
x4

y3 +

y3

3 y
2 y2 2

x6

y6

4 y0 4
5 y0 + 10
4 y1 + 6

3 y
3 y

y4
y5

1
2

+ 3

2 y3 +

y4

x5

4 y2 4

5 y1 10

2 y4
y5

Table 3.3 shows the propagation of error in a dierence table and how the error aects
the dierences. From this table, the following observations are noted.
(i) The eect of the error increases with the order of the dierences.
(ii) The error is maximum (in magnitude) along the horizontal line through the erroneous tabulated value.
(iii) The second dierence column has the errors , 2, , in the third dierence column, the errors are , 3, 3, . In the fourth dierence column the expected
errors , 4, 6, 4, (this column is not sucient to show all of the expected errors). Thus, in the pth dierence column, the coecients of errors are the binomial
coecients in the expansion of (1 x)p .

90 Numerical Analysis
(iv) The algebraic sum of errors in any column (complete) is zero. If there is any error
in a single entry of a table, then from the dierence table one can detect and
correct such error.
Detection of errors using dierence table
Dierence table may be used to detect errors in a set of tabular values. From Table
3.3, it follows that if an error is present in a given data, the dierences of some order
will become alternating in sign. Thus, higher order dierences should be formed till the
error is revealed.
To detect the position of the error in an entry, the following steps may be proceed.
(i) Form the dierence table. If at any stage, the dierences do not follow a smooth
pattern, then one can conclude that there is an error.
(ii) If the dierences of some order (it is generally happens in higher order) becomes
alternating in sign then the middle entry has an error.

3.5

## Newtons Forward Dierence Interpolation Formula

Let y = f (x) be a function whose explicit form is unknown. But, the values of y at
the equispaced points x0 , x1 , . . . , xn , i.e., yi = f (xi ), i = 0, 1, . . . , n are known. Since
x0 , x1 , . . . , xn are equispaced then xi = x0 + ih, i = 0, 1, . . . , n, where h is the spacing.
It is required to construct a polynomial (x) of degree less than or equal to n satisfying
the conditions
yi = (xi ),

i = 0, 1, . . . , n.

(3.32)

Since (x) is a polynomial of degree at most n, so (x) can be taken in the following
form
(x) = a0 + a1 (x x0 ) + a2 (x x0 )(x x1 ) + a3 (x x0 )(x x1 )(x x2 )
+ + an (x x0 )(x x1 ) (x xn1 ),

(3.33)

## where a0 , a1 , . . . , an are constants whose values are to be determined using (3.32).

To determine the values of ai s, substituting x = xi , i = 0, 1, 2, . . . , n.
When x = x0 then
(x0 ) = a0 or, a0 = y0 .
For x = x1 , (x1 ) = a0 + a1 (x1 x0 )
y0
y1 y 0
=
.
or, y1 = y0 + a1 h or, a1 =
h
h

Interpolation

91

## For x = x2 , (x2 ) = a0 + a1 (x2 x1 ) + a2 (x2 x0 )(x2 x1 )

y1 y 0
.2h + a2 (2h)(h)
or, y2 = y0 +
h
y2 2y1 + y0
2 y0
=
.
or, a2 =
2!h2
2!h2
In this way,
3 y0
4 y0
n y0
a3 =
,
a
=
,
.
.
.
,
a
=
.
4
n
3!h3
4!h4
n!hn
Using these values, (3.33) becomes
y0
2 y0
+ (x x0 )(x x1 )
h
2!h2
3
y0
+(x x0 )(x x1 )(x x2 )
3!h3
n y0
+ + (x x0 )(x x1 ) (x xn1 )
.
n!hn

(x) = y0 + (x x0 )

(3.34)

## Introducing the condition xi = x0 + ih, i = 0, 1, . . . , n for equispaced points and a

new variable u as x = x0 + uh.
Therefore, x xi = (u i)h.
So the equation (3.34) becomes
2 y0
y0
3 y0
+ (uh)(u 1)h
+ (uh)(u 1)h(u 2)h
2
h
2!h
3!h3
n
y0
+ + (uh)(u 1)h(u 2)h (u n 1)h
n!hn
u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0
= y0 + uy0 +
2!
3!
u(u 1)(u 2) (u n 1) n
y0 ,
+ +
(3.35)
n!

(x) = y0 + (uh)

x x0
.
where u =
h
This is known as Newton or Newton-Gregory forward dierence interpolating
polynomial.
Example 3.5.1 The following table gives the values of ex for certain equidistant
values of x. Find the value of ex when x = 0.612 using Newtons forward dierence
formulae.
x
y

:
:

0.61
1.840431

0.62
1.858928

0.63
1.877610

0.64
1.896481

0.65
1.915541

92 Numerical Analysis
Solution. The forward dierence table is
x
0.61

y
1.840431

0.62

1.858928

0.63

1.877610

0.64

1.896481

2 y

3 y

0.018497
0.000185
0.018682

0.000004
0.000189

0.018871

0.0
0.000189

0.019060
0.65

1.915541

## Here, x0 = 0.61, x = 0.612, h = 0.01, u =

Then,

0.612 0.61
x x0
=
= 0.2.
h
0.01

u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0
2!
3!
0.2(0.2 1)
0.000185
= 1.840431 + 0.2 0.018497 +
2
0.2(0.2 1)(0.2 2)
0.000004
+
6
= 1.840431 + 0.003699 0.000015 + 0.00000019
= 1.844115.

y(0.612) = y0 + uy0 +

3.5.1

## The error in any polynomial interpolation formula is

E(x) = (x x0 )(x x1 ) (x xn )

f (n+1) ()
(n + 1)!

## = u(u 1)(u 2) (u n)hn+1

f (n+1) ()
(using x = x0 + uh)
(n + 1)!

## where lies between min{x0 , x1 , . . . , xn , x} and max{x0 , x1 , . . . , xn , x}.

Also, f (n+1) ()  hn+1 n+1 y0 .
Therefore,
u(u 1)(u 2) (u n) n+1

y0 .
E(x) 
(n + 1)!

Interpolation

93

A particular case:
If 0 < u < 1 then


2
1
1
1
|u(u 1)| = (1 u)u = u u =
u and
4
2
4
|(u 2)(u 3) (u n)| |(2)(3) (n)| = n!.
Then,
1
1 n!
|n+1 y0 | =
|n+1 y0 |.
|E(x)|
4 (n + 1)!
4(n + 1)
2

## Also, |n+1 y0 | 9 in the last signicant gure.

9
< 1 for n > 2 and 0 < u < 1.
Thus, |E(x)|
4(n + 1)
That is, the maximum error in Newtons forward interpolation is 1 when |x x0 | < h.
Newtons forward formula is used to compute the approximate value of f (x) when
the argument x is near the beginning of the table. But this formula is not appropriate
to compute f (x) when x at the end of the table. In this situation Newtons backward
formula is appropriate.
Algorithm 3.2 (Newtons forward interpolation). This algorithm determines
the value of y when the value of x is given, by Newtons forward interpolation method.
The values of xi , yi , i = 0, 1, 2, . . . , n are given and assumed that xi = x0 + ih,
i.e., the data are equispaced.
Algorithm Newton Forward Intepolation
//Assume that the data are equispaced.//
Read (xi , yi ), i = 0, 1, 2, . . . , n;
Read xg; //the value of x at which y is to be determined.//
Compute h = x1 x0 ; //compute spacing.//
Compute u = (xg x0 )/h;
for j = 0 to n do
dyj = yj ; //copy of y to dy//
Set prod = 1, sum = y0 ;
for i = 1 to n do
for j = 0 to (n i) do dyj = dyj+1 dyj ;
//dy represents the dierence.//
ui+1
;
Compute prod = prod
i
Compute sum = sum + prod dy0 ;
endfor;
Print The value of y at x =,xg, is , sum;
end Newton Forward Intepolation

94 Numerical Analysis

Program 3.2
.
/* Program Newton Forward Interpolation
This program finds the value of y=f(x) at a given x when
the function is supplied as (x[i],y[i]), i=0, 1, ..., n.
Assumed that xs are equispaced.*/
#include<stdio.h>
#include<math.h>
void main()
{
int i,j,n; float x,y,xg,sum,prod=1,u,dy,h;
printf("Enter number of subintervals ");
scanf("%d",&n);
printf("Enter x and y values ");
for(i=0;i<=n;i++) scanf("%f %f",&x[i],&y[i]);
printf("Enter interpolating point x ");
scanf("%f",&xg);
h=x-x;
u=(xg-x)/h;
for(j=0;j<=n;j++) dy[j]=y[j];
prod=1; sum=y;
for(i=1;i<=n;i++)
{
for(j=0;j<=n-i;j++) dy[j]=dy[j+1]-dy[j];
prod*=(u-i+1)/i;
sum+=prod*dy;
}
printf("The value of y at x=%f is %f ",xg,sum);
}
A sample of input/output:
Enter number of subintervals 4
Enter x and y values
140 3.685
150 5.854
160 6.302
170 8.072
180 10.225
Enter interpolating point x 142
The value of y at x=142.000000 is 4.537069

Interpolation

3.6

95

## Suppose, a set of values y0 , y1 , . . . , yn of the function y = f (x) is given, at x0 , x1 , . . . , xn ,

i.e., yi = f (xi ), i = 0, 1, . . . , n. Let xi s are equispaced with spacing h, i.e., xi = x0 + ih.
Let us consider the polynomial (x) in the following form:
f (x)  (x) = a0 + a1 (x xn ) + a2 (x xn )(x xn1 )
+a3 (x xn )(x xn1 )(x xn2 )
+ + an (x xn )(x xn1 ) (x x1 ).

(3.36)

## The constants ai s are to be determined using the conditions

yi = (xi ), i = 0, 1, . . . , n.
Substituting x = xn , nn1 , . . . , x1 in (3.36), we obtain
(xn ) = a0 or, a0 = yn .
(xn1 ) = a0 + a1 (xn1 xn ) or, yn1 = yn + a1 (h) or, a1 =

(3.37)

yn yn1
yn
=
.
h
h

## (xn2 ) = a0 + a1 (xn2 xn ) + a2 (xn2 xn )(xn2 xn1 )

yn yn1
(2h) + a2 (2h)(h)
= yn +
h
yn 2yn1 + yn2
2 yn
=
.
or, yn2 = 2yn1 yn + a2 .2!h2 or, a2 =
2!h2
2!h2
In this way, the others values are obtained as,
a3 =

3 yn
4 yn
n yn
,
a
=
,
.
.
.
,
a
=
.
4
n
3!h3
4!h4
n!hn

When the values of ai s are substituted in (3.36) then the polynomial (x) becomes
yn
2 yn
+ (x xn )(x xn1 )
h
2!h2
3
yn
+(x xn )(x xn1 )(x xn2 )
+
3!h3
n yn
.
+(x xn )(x xn1 )(x xn2 ) (x x1 )
n!hn

(x) = yn + (x xn )

(3.38)

## Now, a unit less variable v is introduced which is dened as x = xn + vh, i.e.,

x xn
v=
. This substitution simplies the formula.
h
Also, for equispaced points xi = x0 + ih.
Then x xni = (xn + vh) (x0 + n ih) = (xn x0 ) + (v n i)h = (v + i)h,
i = 0, 1, . . . , n.

96 Numerical Analysis
Using above results, (3.38) becomes
2 yn
yn
3 yn
+ vh(v + 1)h
+
vh(v
+
1)h(v
+
2)h
+
h
2!h2
3!h3
n yn
+vh(v + 1)h(v + 2)h (v + n 1)h
n!hn
v(v + 1) 2
v(v + 1)(v + 2) 3
yn +
yn +
= yn + vyn +
2!
3!
v(v + 1)(v + 2) (v + n 1) n
yn .
(3.39)
+
n!

(x) = yn + vh

## This formula is known as Newtons backward or Newton-Gregory backward

interpolation formula.
Example 3.6.1 From the following table of values of x and f (x) determine the
value of f (0.29) using Newtons backward interpolation formula.
x
f (x)

:
:

0.20
1.6596

0.22
1.6698

0.24
1.6804

0.26
1.6912

0.28
1.7024

0.30
1.7139

x
0.20
0.22
0.24
0.26
0.28
0.30

f (x)
1.6596
1.6698
1.6804
1.6912
1.7024
1.7139

f (x)

2 f (x)

3 f (x)

0.0102
0.0106
0.0108
0.0112
0.0115

0.0004
0.0002
0.0004
0.0003

0.0002
0.0002
0.0001

## Here, xn = 0.30, x = 0.29, h = 0.02, v =

Then,

0.29 0.30
x xn
=
= 0.5.
h
0.02

v(v + 1) 2
v(v + 1)(v + 2) 3
f (xn )+
f (xn )+
2!
3!
0.5(0.5 + 1)
0.0003
= 1.7139 0.5 0.0115 +
2
0.5(0.5 + 1)(0.5 + 2)
(0.0001)
+
6
= 1.7139 0.00575 0.0000375 + 0.00000625
= 1.70811875  1.7081.

## f (0.29) = f (xn ) + vf (xn )+

Interpolation

97

Example 3.6.2 The population of a town in decennial census were given in the
following table.
Year
Population (in thousand)

:
:

1921
46

1931
66

1941
81

1951
93

1961
101

Estimate the population for the year 1955 using Newtons backward and forward
formulae and compare the results.
Solution.
Using Newtons backward formula
The backward dierence table is
Population
(y)
46
66
81
93
101

2 y

3 y

4 y

20
15
12
8

5
3
4

2
1

## Here, xn = 1961, x = 1955, h = 10, v =

By Newtons backward formula

xxn
h

Year
(x)
1921
1931
1941
1951
1961

19551961
10

= 0.6.

v(v + 1) 2
v(v + 1)(v + 2) 3
yn +
yn
2!
3!
v(v + 1)(v + 2)(v + 3) 4
yn
+
4!
0.6(0.6 + 1)
(4)
= 101 0.6 8 +
2
0.6(0.6 + 1)(0.6 + 2)
(1)
+
6
0.6(0.6 + 1)(0.6 + 2)(0.6 + 3)
(3)
+
24
= 96.8368  97.

y(1955) = yn + vyn +

## Hence the approximate population of the town was 97 thousand.

Using Newtons forward formula
The given table is written in reverse order as
Year
Population

:
:

1961
101

1951
93

1941
81

1931
66

1921
46

98 Numerical Analysis
The forward dierence table is
Year
x
1961

Population
y
101

1951

93

y
8
12

1941

81
15

1931

66
20

1921

2 y

3 y

4 y

4
1
3
5

46

## Here x0 = 1961, x = 1955, h = 10, u =

Then

1955 1961
x x0
=
= 0.6.
h
10

u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0
2!
3!
u(u 1)(u 2)(u 3) 4
y0
+
4!
0.6(0.6 1)(0.6 2)
0.6(0.6 1)
(4) +
1
= 101 + 0.6 (8) +
2
6
0.6(0.6 1)(0.6 2)(0.6 3)
(3)
+
24
= 101 4.8 + 0.48 + 0.056 + 0.1008
= 96.8368  97.

y(1955) = y0 + uy0 +

Therefore the population of the town in the year 1955 was 97 thousand and this result
is same with the result obtained by Newtons backward dierence formula.
3.6.1

## The error in this interpolation formula is

E(x) = (x xn )(x xn1 ) (x x1 )(x x0 )
= v(v + 1)(v + 2) (v + n)hn+1
where v =

f (n+1) ()
(n + 1)!

f (n+1) ()
,
(n + 1)!

(3.40)

x xn
and lies between min{x0 , x1 , . . . , xn , x} and max{x0 , x1 , . . . , xn , x}.
h

Interpolation

99

Note 3.6.1 The Newtons backward dierence interpolation formula is used to compute
the value of f (x) when x is near to xn , i.e., when x is at the end of the table.

3.7

## Gaussian Interpolation Formulae

Newtons forward and Newtons backward formulae does not give accurate value of f (x)
when x is in the middle of the table. To get more accurate result another formula may
be used. There are several methods available to solve this type of problem. Among
them Gaussian forward and backward, Stirlings and Bessels interpolation formulae are
widely used.
3.7.1

## Case I: For 2n + 1 (odd) arguments

Suppose the values of the function y = f (x) are known at 2n + 1 equally spaced points
xn , x(n1) , . . . , x1 , x0 , x1 , . . . , xn1 , xn , i.e., yi = f (xi ), i = 0, 1, 2, . . . , n.
The problem is to construct a polynomial (x) of degree at most 2n such that
(xi ) = yi , i = 0, 1, 2, . . . , n,

(3.41)

## where xi = x0 + ih, h is the spacing.

Let us consider (x) as
(x) = a0 + a1 (x x0 ) + a2 (x x0 )(x x1 ) + a3 (x x1 )(x x0 )(x x1 )
+a4 (x x1 )(x x0 )(x x1 )(x x2 ) +
+a2n1 (x xn+1 )(x xn+2 ) (x x1 )(x x0 ) (x xn1 )
+a2n (x xn+1 )(x xn+2 ) (x x1 )(x x0 ) (x xn ),
(3.42)
where ai s are unknown constants and their values are to be determined by substituting
x = x0 , x1 , x1 , x2 , x2 , . . . , xn , xn .
Therefore,
y0 = a0
y1 = a0 + a1 (x1 x0 ) i.e., y1 = y0 + a1 h,
y1 y 0
y0
i.e., a1 =
.
=
x1 x0
h
y1 = y0 + a1 (h) + a2 (h)(2h)
y0
= y0 h
+ a2 h2 2!
h

## 100 Numerical Analysis

y1 2y0 + y1
2 y1
=
.
2! h2
2! h2
y2 = a0 + a1 (x2 x0 ) + a2 (x2 x0 )(x2 x1 )

i.e., a2 =

## +a3 (x2 x1 )(x2 x0 )(x2 x1 )

y1 2y0 + y1
y1 y 0
(2h) +
(2h)(h) + a3 (3h)(2h)(h)
= y0 +
h
2!h2
y2 3y1 + 3y0 y1
3 y1
or, a3 =
=
.
3!h3
3!h3
In this manner, the remaining values are obtained as
2n1 y(n1)
4 y2
5 y2
2n yn
a4 =
,
a
=
,
.
.
.
,
a
=
,
a
=
.
5
2n1
2n
4!h4
5!h5
(2n 1)!h2n1
(2n)!h2n
Thus the Gausss forward dierence formula is
y0
2 y1
+ (x x0 )(x x1 )
h
2!h2
3
y1
+(x x1 )(x x0 )(x x1 )
+
3!h3
2n1 y(n1)
+(x x(n+1) ) (x xn1 )
(2n 1)!h2n1
2n yn
+(x x(n+1) ) (x xn1 )(x xn )
.
(2n)!h2n

(x) = y0 + (x x0 )

## To simplify the above equation, a new variable s is introduced, where

x x0
s=
i.e, x = x0 + sh.
h
Also, x i = x0 ih, i = 0, 1, 2, . . . , n.
Therefore, x xi = (s i)h.
Then x x0 = sh, x x1 = (s 1)h, x x1 = (s + 1)h,
x x2 = (s 2)h, x x2 = (s + 2)h and so on.
Making use of these results, (3.43) becomes
2 y1
y0
3 y1
+ sh(s 1)h
+
(s
+
1)hsh(s

1)h
+
h
2!h2
3!h3
2n1 y(n1)
+(s + n 1)h sh(s 1)h (s n 1)h
(2n 1)!h2n1
2n yn
+(s + n 1)h sh(s 1)h (s n 1)h(s n)h
(2n)!h2n

(x) = y0 + sh

(3.43)

Interpolation
3 y1
2 y1
+ (s + 1)s(s 1)
+
2!
3!
2n1 y(n1)
+(s + n 1) s(s 1) (s n 1)
(2n 1)!
2n yn
+(s + n 1) s(s 1) (s n 1)(s n)
(2n)!
2
3
y1
y1
+ s(s2 12 )
= y0 + sy0 + s(s 1)
2!
3!
4y

2
+
+s(s2 12 )(s 2)
4!
2n1 y(n1)
2
2
+s(s2 n 1 )(s2 n 2 ) (s2 12 )
(2n 1)!
2n yn
2
2
.
+s(s2 n 1 )(s2 n 2 ) (s2 12 )(s n)
(2n)!

101

= y0 + sy0 + s(s 1)

(3.44)

## The formula (3.43) or (3.44) is known as Gausss forward central dierence

formula or the rst interpolation formula of Gauss.

## Case II: For 2n (even) arguments

In this case the arguments are x0 , x1 , . . . , x(n1) and xn .
For these points the Gausss forward interpolation takes the following form.
2 y1
3 y1
+ (s + 1)s(s 1)
2!
3!
4
y2
+(s + 1)s(s 1)(s 2)
4!
5 y2
+
+(s + 2)(s + 1)s(s 1)(s 2)
5!
2n1 y(n1)
+(s + n 1) s (s n 1)
(2n 1)!
2
y1
3 y1
+ s(s2 12 )
= y0 + sy0 + s(s 1)
2!
3!
4
y2
5 y2
+s(s2 12 )(s 2)
+ (s2 22 )(s2 12 )s
+
4!
5!
2n1 y(n1)
2
.
+(s2 n 1 ) (s2 12 )s
(2n 1)!

(3.45)

3.7.2

## The remainder of Gausss forward central dierence interpolation for 2n + 1 arguments

is
f 2n+1 ()
(2n + 1)!
= (s + n)(s + n 1) (s + 1)s(s 1) (s n + 1)(s n)
f 2n+1 ()
h2n+1
(2n + 1)!
2n+1
f
()
(3.46)
= s(s2 12 ) (s2 n2 ).h2n+1
(2n + 1)!

## E(x) = (x xn )(x x(n1) ) (x x1 )(x x0 ) (x xn )

x x0
and lies between min{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 , xn }
h
and max{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 , xn }.
In case of 2n arguments, the error is

where s =

2

f 2n ()
,
(2n)!

f 2n ()
(2n)!
(3.47)

## where, min{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 , xn } <

< max{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 , xn }.
3.7.3

## Case I: For 2n + 1 (odd) number of arguments

Let a function y = f (x) is known for 2n + 1 equispaced arguments xi , i = 0, 1, 2, . . . , n,
such that
xi = x0 ih, i = 0, 1, 2, . . . , n.
Let yi = f (xi ), i = 0, 1, 2, . . . , n.
Our aim is to determine a polynomial (x) of degree not more than 2n which satises
the conditions
(xi ) = yi , i = 0, 1, . . . , n.

(3.48)

Interpolation

103

## The polynomial (x) is considered in the following form.

(x) = a0 + a1 (x x0 ) + a2 (x x1 )(x x0 ) + a3 (x x1 )(x x0 )(x x1 )
+a4 (x x2 )(x x1 )(x x0 )(x x1 )
+a5 (x x2 )(x x1 )(x x0 )(x x1 )(x x2 ) +
+a2n1 (x x(n1) ) (x x1 )(x x0 ) (x xn1 )
+a2n (x xn )(x x(n1) ) (x x1 )(x x0 ) (x xn1 ).

(3.49)

The coecients ai s are unknown constants. These values are determined by using
the relations (3.48). Substituting x = x0 , x1 , x1 , x2 , x2 , . . . , xn , xn to (3.49) in succession. Note that xi xj = (i + j)h and (xi xj ) = (i + j)h. Then it is found
that
y0 = a0
(x1 ) = a0 + a1 (x1 x0 )
i.e., y1 = y0 + a1 (h),
y0 y1
y1
i.e., a1 =
=
h
h
(x1 ) = a0 + a1 (x1 x0 ) + a2 (x1 x1 )(x1 x0 )
y1
+ a2 (2h)(h)
y1 = y0 + h.
h
y1 y0 (y0 y1 )
2 y1
i.e., a2 =
=
2!h2
2!h2
(x2 ) = a0 + a1 (x2 x0 ) + a2 (x2 x1 )(x2 x0 )
+a3 (x2 x1 )(x2 x0 )(x2 x1 )
2 y1
y1
(2h) +
i.e., y2 = y0 +
(h)(2h) + a3 (h)(2h)(3h)
h
2!h2
= y0 2(y0 y1 ) + (y1 2y0 + y1 ) + a3 (1)3 (3!)h3
y1 3y0 + 3y1 y2
3 y2
or, a3 =
=
.
3!h3
3!h3
In this manner, the other values are obtained as
a4 =

4 y2
5 y3
2n1 yn
2n yn
,
a
=
,
.
.
.
,
a
=
,
a
=
.
5
2n1
2n
4!h4
5!h5
(2n 1)!h2n1
(2n)!h2n

## 104 Numerical Analysis

y1
2 y1
+ (x x1 )(x x0 )
1!h
2!h2
3
y2
+(x x1 )(x x0 )(x x1 )
3!h3
4 y2
+(x x2 )(x x1 )(x x0 )(x x1 )
+
4!h4

(x) = y0 + (x x0 )

2n1 yn
(2n 1)!h2n1
2n yn
+(x xn )(x x1 )(x x0 )(x x1 ) (x xn1 )
.
(3.50)
(2n)!h2n

## +(x x(n1) ) (x x1 )(x x0 ) (x xn1 )

As in previous case, a new unit less variable s is introduced to reduce the above
x x0
i.e, x = x0 + sh.
formula into a simple form, where s =
h
Then
x x0 ih
x xi
=
= s i and
h
h
x x0 + ih
x xi
=
= s + i, i = 0, 1, 2, . . . , n.
h
h
Then the above formula is transferred to
(s + 1)s 2
(s + 1)s(s 1) 3
y1 +
y2
2!
3!
(s + 2)(s + 1)s(s 1) 4
y2 +
+
4!
(s + n 1) (s + 1)s(s 1) (s n + 1) 2n1

yn
+
(2n 1)!
(s + n)(s + n 1) (s + 1)s(s 1) (s n + 1) 2n
yn .
+
(2n)!

(x) = y0 + sy1 +

(3.51)

## The above formula (3.51) is known as Gausss backward interpolation formula

or second interpolation formula of Gauss.
Case II: For 2n (even) number of arguments
In this case the arguments are taken as x0 , x1 , . . . , x(n1) and xn , where xi =
x0 ih, i = 0, 1, . . . , n 1 and xn = x0 nh.

Interpolation

105

## For these equispaced points the Gausss backward interpolation formula is

(s + 1)s 2
(s + 1)s(s 1) 3
y1 +
y2
2!
3!
(s + 2)(s + 1)s(s 1) 4
y2 +
+
4!
(s + n 1) (s + 1)s(s 1) (s n + 1) 2n1

yn ,
+
(2n 1)!

(x) = y0 + sy1 +

where s =

3.7.4

(3.52)

x x0
.
h

## The remainder for (2n + 1) equispaced points is

f 2n+1 ()
(2n + 1)!
= (s + n)(s + n 1) (s + 1)s(s 1) (s n + 1)(s n)
f 2n+1 ()
,
h2n+1
(2n + 1)!

## where min{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 , xn } <

< max{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 , xn }.
The remainder for the case of 2n equispaced points is
f 2n ()
(2n)!
2n
f ()
,
= (s + n)(s + n 1) (s + 1)s(s 1) (s n + 1)h2n
(2n)!

## min{xn , xn , . . . , x0 , x1 , . . . , xn1 , xn1 } <

< max{xn , x(n1) , . . . , x0 , x1 , . . . , xn1 }.

3.8

## Stirlings interpolation formula is used for odd number of equispaced arguments.

This formula is obtained by taking the arithmetic mean of the Gausss forward and
backward dierence formulae given by (3.44) and (3.51).

## 106 Numerical Analysis

Therefore Stirlings formula is
(x)forward + (x)backward
2
s y1 + y0 s2 2
s(s2 12 ) 3 y2 + 3 y1
+ y1 +
= y0 +
1!
2
2!
3!
2
s2 (s2 12 ) 4
s(s2 12 )(s2 22 ) 5 y3 + 5 y2
+
y2 +
+
4!
5!
2
2
s2 (s2 12 )(s2 22 ) (s2 n 1 ) 2n
(3.53)
yn .
+
(2n)!

(x) =

E(x) =

h
f
(),
(2n + 1)

(3.54)

## where min{xn , . . . , x0 , . . . , xn } < < max{xn , . . . , x0 , . . . , xn }.

The formula (3.53) is known as Stirlings central dierence interpolation formula.
Note 3.8.1 (a) The Stirlings interpolation formula (3.53) gives the best approximate
x x0
result when 0.25 < s < 0.25. So we choose x0 in such a way that s =
satisfy
h
this condition.
(b) The Stirlings interpolation formula is used when the point x, for which f (x) to
be determined, is at the centre of the table and the number of points at which the values
of f (x) known is odd.

3.9

## Bessels Interpolation Formula

This central dierence formula is also obtained by taking the arithmetic mean of Gausss
forward and backward interpolation formulae. But, one dierence is that the backward
formula taken after one modication.
Let us consider 2n equispaced points x(n1) , . . . , x1 , x0 , x1 , . . . , xn1 , xn as arguments, where xi = x0 ih, h is the spacing.
If x0 , y0 be the initial values of x and y respectively, then the Gausss backward
dierence interpolation formula (3.52) is
s(s + 1) 2
(s + 1)s(s 1) 3
y1 +
y2
2!
3!
(s + 2)(s + 1)s(s 1) 4
y2 +
+
4!
(s + n 1) (s + 1)s(s 1) (s n + 1) 2n1

yn .
+
(2n 1)!

(x) = y0 + sy1 +

(3.55)

Interpolation

107

## Suppose x1 , y1 be the initial values of x and y. Then

x (x0 + h)
x x0
x x1
=
=
1 = s 1.
h
h
h
Also, the indices of all the dierences of (3.55) will increase by 1. Now, replacing s
by s 1 and increasing the indices of (3.55) by 1, then the above equation becomes
s(s 1) 2
s(s 1)(s 2) 3
y0 +
y1
2!
3!
(s + 1)s(s 1)(s 2) 4
y1
+
4!
(s + 1)s(s 1)(s 2)(s 3) 5
y2 +
+
5!
(s + n 2) (s + 1)s(s 1)(s 2) (s n) 2n1

yn+1 . (3.56)
+
(2n 1)!

1 (x) = y1 + (s 1)y0 +

Taking arithmetic mean of (3.56) and Gausss forward interpolation formula given by
(3.45),
1 (x) + (x)forward
2
y 0 + y1 
1
s(s 1) 2 y0 + 2 y1
=
+ s
y0 +
2
2
2!
2
1
(s 2 )s(s 1) 3
s(s 1)(s + 1)(s 2) 4 y2 + 4 y1
+
y1 +
3!
4!
2
1
(s 2 )s(s 1)(s + 1)(s 2) 5
y2 +
+
5!
(s 12 )s(s 1)(s + 1) (s + n 2)(s n 1) 2n1

+
y(n1) , (3.57)
(2n 1)!

(x) =

x x0
.
h
x x0 1
1
and then the above formula reduces to
Introducing u = s =
2
h
2

where s =

(x) =

u2 14 2 y1 + 2 y0 u(u2 14 ) 3
y0 + y 1
+ uy0 +

+
y1
2
2!
2
3!
2
(u2 14 )(u2 94 ) (u2 (2n3)
) 2n1
4

+
y(n1) .
(2n 1)!

(3.58)

The formulae given by (3.57) and (3.58) are known as Bessels central dierence
interpolation formula.

## 108 Numerical Analysis

Note 3.9.1 (a) The Bessels formula gives the best result when the starting point x0
be so chosen such that 0.25 < u < 0.25 i.e., 0.25 < s0.5 < 0.25 or, 0.25 < s < 0.75.
(b) This central dierence formula is used when the interpolating point is near the
middle of the table and the number of arguments is even.

Example 3.9.1 Use the central dierence interpolation formula of Stirling or Bessel
to nd the values of y at (i) x = 1.40 and (ii) x = 1.60 from the following table
x
y

:
:

1.0
1.0000

1.25
1.0772

1.50
1.1447

1.75
1.2051

2.00
2.2599

i
2

xi
1.00

yi
1.0000

1.25

1.0772

1.50

1.1447

1.75

1.2051

yi

2 yi

3 yi

0.0772
0.0097
0.0675

0.0026
0.0071

0.0604

0.0015
0.0056

0.0548
2

2.00

1.2599

## (i) For x = 1.40, we take x0 = 1.50, then s = (1.40 1.50)/0.25 = 0.4.

The Bessels formula gives
1
s(s 1) 2 y0 + 2 y1
y0 + y1 
+ s
y0 +
2
2
2!
2
1
1
3
s
s(s 1) y1
+
3!
2
1.1447 + 1.2051
+ (0.4 0.5) 0.0604
=
2
0.4(0.4 1) 0.0071 0.0056
+
2!
2
1
+ (0.4 0.5)(0.4)(0.4 1) 0.0015
6
= 1.118636.

y(1.40) =

Interpolation

109

## Using Stirlings formula

y1 + y0 s2 2
s(s2 12 ) 3 y2 + 3 y1
+ y1 +
2
2!
3!
2
2
0.0675 + 0.0604 (0.4)
+
(0.0071)
= 1.1447 + 0.4
2
2
0.4(0.16 1) 0.0026 + 0.0015
+
6
2
= 1.1447 + 0.02558 0.000568 0.0001148 = 1.1695972.

y(1.60) = y0 + s

3.10

## In this interpolation formula, 2n (even) equispaced arguments x(n1) , x(n2) , . . .,

x1 , x0 , x1 , . . . , xn1 and xn are considered. The Everetts interpolation formula is
obtained from Gausss forward interpolation formula (3.44) by replacing the odd order
dierences using the lower order even dierences. That is, by the substitution
2k+1 yk = 2k y(k1) 2k yk .
That is,
y0 = y1 y0 ;
y1 = 2 y0 2 y1 ;
3

5 y2
..
.

= 4 y1 4 y2 ,
..
..
.
.

## 2n1 y(n1) = 2n2 y(n2) 2n2 y(n1) .

On substitution of these relations, equation (3.44) yields
s(s 1) 2
(s + 1)s(s 1) 2
y1 +
( y0 2 y1 )
2!
3!
(s + 1)s(s 1)(s 2) 4
y2
+
4!
(s + 2)(s + 1)s(s 1)(s 2) 4
( y1 4 y2 ) +
+
5!
(s + n 1)(s + n 2) (s + 1)s(s 1)(s n + 1)
+
(2n 1)!
2n2
(
y(n2) 2n2 y(n1) )

(s + 1)s(s 1) 2
(s + 2)(s + 1)s(s 1)(s 2) 4
y0 +
y1 +
= sy1 +
3!
5!

(x) = y0 + s(y1 y0 ) +

## (s + n 1)(s + n 2) (s + 1)s(s 1) (s n + 1) 2n2

+
y(n2)
(2n 1)!

 s(s 1) (s + 1)s(s 1) 
+ (1 s)y0 +

2 y1
2!
3!
 (s + 1)s(s 1)(s 2) (s + 2)(s + 1)s(s 1)(s 2) 

4 y2 +
+
4!
5!

## (s + n 1)(s + n 2) (s + 1)s(s 1) (s n + 1) 2n2

y(n1)
+
(2n 1)!

s(s2 12 ) 2
s(s2 11 )(s2 22 ) 4
y0 +
y1 +
= sy1 +
3!
5!

## s(s2 12 )(s2 22 ) (s2 (n 1)2 ) 2n2

+
y(n2)
(2n 1)!

u(u2 11 ) 2
u(u2 12 )(u2 22 ) 4
y1 +
y2 +
+ uy0 +
3!
5!

+

y(n1) ,
(3.59)
(2n 1)!

where s =

x x0
and u = 1 s.
h

## Example 3.10.1 Use Everetts interpolation formula to nd the value of y when

x = 1.60 from the following table.
x
y

:
:

1.0
1.0000

1.25
1.1180

1.50
1.2247

1.75
1.3229

2.00
1.4142

2.25
1.5000

i
2

xi
1.00

yi
1.0000

1.25

1.1180

yi

2 yi

3 yi

4 yi

0.1180
0.0113
0.1067
0

1.50

0.0028
0.0085

1.2247
0.0982

1.75

0.0069

1.3229
0.0913

2.00
2.25

1.5000

0.0002
0.0014

0.0055

1.4142
0.0858

0.0012
0.0016

Interpolation

111

## We take x0 = 1.50. Here h = 0.25.

1.60 1.50
x x0
=
= 0.4 and u = 1 s = 0.6.
Then s =
h
0.25
Using Everetts formula, the value of y(1.60) is given by

s(s2 12 ) 2
s(s2 12 )(s2 22 ) 4
y(1.60) = sy1 +
y0 +
y1
3!
5!

u(u2 12 ) 2
u(u2 12 )(u2 22 ) 4
y1 +
y2
+ uy0 +
3!
5!

0.4(0.16 1)
= 0.4 1.3229 +
(0.0069)
6

0.4(0.16 1)(0.16 4)
(0.0002) + 0.6 1.2247
+
120

0.6(0.36 1)(0.36 4)
0.6(0.36 1)
(0.0085) +
(0.0012)
+
6
120
= 0.5292 + 0.0004 0.0000025 + 0.7348 + 0.0005 0.00001
= 1.2649.

3.10.1

## Relation between Bessels and Everetts formulae

These two formulae are closely related and one formula can be deduced from the other
one. Now, starting from Bessels formula (3.57):
y 0 + y1 
1
s(s 1) 2 y0 + 2 y1
+ s
y0 +
2
2!
2
2

1
s 2 s(s 1)
x x0
3 y1 + , where s =
.
+
3!
h
y 0 + y1 
1
s(s 1) 2 y0 + 2 y1
=
+ s
(y1 y0 ) +
2
2!
2

2
1
s 2 s(s 1)
(2 y0 2 y1 ) +
+
3!

s(s 1) s(s 1)(s 12 ) 2

y1 +
= (1 s)y0 +
4
6

s(s 1) s(s 1)(s 12 ) 2
+sy1 +
+
y0 +
4
6
s(s2 12 ) 2
u(u2 12 ) 2
= sy1 +
y0 + + uy0 +
y1 +
3!
3!
where u = 1 s.

(x) =

## 112 Numerical Analysis

This is the Everetts formula up to second order dierences. From this deduction it is
clear that the Everetts formula truncated after second order dierences is equivalent to
the Bessels formula truncated after third dierences. Conversely, the Bessels formula
may be deduced from Everetts formula.

## Central dierence table

The dierence Table 3.3 shows how the dierent order of dierences are used in dierent
interpolation formulae.
x

x3

y3

2 y

3 y

4 y

y3
x2

y2

2 y


3

33

y2
x1

y1

Newtons Backward

2 y2

4 y3

3 y2

y1

x0

## y0 - 6-2 y1- 6-4 y2- Stirlings formula

?

x1

y1

?
R
- 3 - 6 - Bessels Formula
6- y0 - 6
? y1
?
?
R
R 3
y0

y1
x2

y2

y3

2 y1

4 y0

3 y1

y2
x3

4 y1

2 y0

Newtons Forward

2 y2

## Figure 3.3: Central dierence table.

Note 3.10.1 In Newtons forward and backward interpolation formulae the rst or the

Interpolation

113

last interpolating point is taken as initial point x0 . But, in central dierence interpolation formulae, a middle point is taken as the initial point x0 .

3.11

## Interpolation by Iteration (Aitkens Interpolation)

Newtons interpolation formula generates successively higher order interpolation formula. The Aitken interpolation formula served the same purpose. But it has one
advantage that it can be easily programmed for a computer.
Let y = f (x) be given for n + 1 distinct points x0 , x1 , . . . , xn , i.e., yi = f (xi ), i =
0, 1, . . . , n are given, where the points xi , i = 0, 1, . . . , n need not be equispaced. To
compute the value of y for a given x the iterations proceed as follows:
to nd the value of y obtain a rst approximation by taking rst two points; then obtain
its second approximation by taking the rst approximations and so on.
The linear polynomial for the points x0 and x1 is
x x1
x x0
1
y0 +
y1 =
[(x1 x)y0 (x0 x)y1 ]
x0 x1
x1 x0
x1 x0


 y0 x0 x 
1
.

=
x1 x0  y1 x1 x 

p01 (x) =

(3.60)

In general,
1
p0j (x) =
xj x0



 y0 x0 x 


 yj xj x  , j = 1, 2, . . . , n.

(3.61)

Here p0j (x) is a polynomial of degree less than or equal to 1, for the points x0 and
xj .
The polynomial


 p01 (x) x1 x 
1
 , j = 2, 3, . . . , n.

p01j (x) =
(3.62)
xj x1  p0j (x) xj x 
is a polynomial of degree less than or equal to 2.
The polynomial p01j (x) interpolates the points x0 , x1 and xj .
In general, the polynomial for the (k + 1) points x0 , x1 , . . . , xk and xj is


 p012k (x) xk x 
1
 , j = k + 1, . . . , n.

p012kj (x) =
xj xk  p012(k1)j (x) xj x 

(3.63)

## This polynomial is of degree (k + 1) and interpolates the points x0 , x1 , . . . , xk1 , xk

and xj .
The tabular representation of this form is

## 114 Numerical Analysis

xj
x0
x1
x2
x3
x4

xn

yj
y0
y1
y2
y3
y4

yn

p0j

p01j

p012j

p0123j

p01
p02
p03
p04

p0n

p012
p013
p014

p01n

p0123
p0124

p012n

p01234

p0123n

xj x
x0 x
x1 x
x2 x
x3 x
x4 x

xn x

Example 3.11.1 Find the value of y(1.52) by iterated linear interpolation using
the following table.
x
y(x)

:
:

1.4
1.8330

1.5
1.9365

1.6
2.0396

p0j

p01j

p012j

1.9572
1.9570
1.9568

1.9572
1.9572

1.9572

1.7
2.1424

## Solution. Here x = 1.52.

The calculations are shown below.
xj
1.4
1.5
1.6
1.7

yj
1.8330
1.9365
2.0396
2.1424

Now
p0j =
p01 =
p02 =
p03 =
p01j =
p012 =
p013 =

xj x
0.12
0.02
0.08
0.18



 y0 x0 x 
1
,

j = 1, 2, 3.
xj x0  yj xj x 


1  1.8330 0.12 
= 1.9572.
0.1  1.9365 0.02 


1  1.8330 0.12 
= 1.9570.
0.2  2.0396 0.08 


1  1.8330 0.12 
= 1.9568.
0.3  2.1424 0.18 


 p01 x1 x 
1

,
j = 2, 3.
xj x1  p0j xj x 


1  1.9572 0.02 
= 1.9572.
0.1  1.9570 0.08 


1  1.9572 0.02 
= 1.9572.
0.2  1.9568 0.18 

Interpolation

p012j
p0123

115



 p012 x2 x 
1
,

=
j = 3.
xj x2  p01j xj x 


1  1.9572 0.08 
=
= 1.9572.
0.1  1.9572 0.18 

## Therefore the interpolated value of y at x = 1.52 is 1.9572.

Algorithm 3.3 (Aitken Interpolation). This algorithm determines the value of
y at a given x from the table of (xi , yi ), i = 0, 1, 2, . . . , n, by Aitkens iterative interpolation formula.
Algorithm Aitken Interpolation
Step 1: Read n, x0 , x1 , . . . , xn ; y0 , y1 , . . . , yn and xg;
// n represents the number of points, xg is the value of
x at which y is to be calculated.//
Step 2: for j = 0 to n do
Step 2.1. Set p(j, 0) = y(j);
Step 2.2. Compute xd(j) = x(j) xg;
endfor;
Step 3: for k = 0 to n do
for j = k + 1 to n do
p(j, k + 1) = [p(k, k) xd(j) p(j, k) xd(k)]/[x(j) x(k)];
endfor;
Step 4: // printing of the table//
for j = 0 to n do
Print x(j), (p(j, k), k = 0 to j), xd(j);
endfor;
Step 5: //Printing of the value of the polynomial.//
Print p(n, n);
end Aitken Interpolation
Program 3.3
.
/* Program Aitken Interpolation
This program implements Aitkens interpolation
formula; xg is the interpolating points. */
#include <stdio.h>
#include <math.h>
void main()
{
int n, j, k;
float x, y, xg, p, xd;

## printf("Enter the value of n and the data points

in the form x[i],y[i] ");
scanf("%d", &n);
for(j=0;j<=n;j++) scanf("%f %f", &x[j],&y[j]);
printf("\nEnter the value of x ");
scanf("%f",&xg);
for(j=0;j<=n;j++){
p[j]=y[j]; xd[j]=x[j]-xg;
}
for(k=0;k<=n;k++)
for(j=k+1;j<=n;j++)
p[j][k+1]=(p[k][k]*xd[j]-p[j][k]*xd[k])/(x[j]-x[k]);
for(j=0;j<=n;j++){
printf("%6.4f ",x[j]);
for(k=0;k<=j;k++) printf(" %6.4f",p[j][k]);
for(k=j+1;k<=n;k++) printf("
");
printf(" %6.4f\n",xd[j]);
}
printf("The value of y at x= %6.3f is %8.5f ",xg,p[n][n]);
} /* main */
A sample of input/output:
Enter the value of n and the data points in the form x[i],y[i] 4
1921 46
1931 68
1941 83
1951 95
1961 105
Enter the value of x 1955
1921.0000 46.0000
-34.0000
1931.0000 68.0000 120.8000
-24.0000
1941.0000 83.0000 108.9000 92.2400
-14.0000
1951.0000 95.0000 101.5333 97.6800 99.8560
-4.0000
1961.0000 105.0000 96.1500 101.0800 98.4280 99.2848 6.0000
The value of y at x= 1955.000 is 99.28481

3.12

## The Lagranges interpolation formula has a disadvantage that if a new interpolation

point is added or removed then the Lagrangian functions Li (x) will have to be re-

Interpolation

117

## computed. But the Newtons general interpolation formula, based on divided

dierences removes this drawback.
Let yi = f (xi ), i = 0, 1, . . . , n be known at n + 1 points x0 , x1 , . . . , xn . The points
x0 , x1 , . . . , xn are not necessarily be equispaced. Then the divided dierences of dierent
orders are dened as follows:
Zeroth order divided dierence
f [x0 ] = f (x0 ).
First order divided dierence
f (x0 ) f (x1 )
.
x0 x1
f (xi ) f (xj )
.
In general, f [xi , xj ] =
xi xj
Second order divided dierence
f [x0 , x1 ] =

f [x0 , x1 ] f [x1 , x2 ]
.
x0 x2
f [xi , xj ] f [xj , xk ]
In general, f [xi , xj , xk ] =
.
xi xk
f [x0 , x1 , x2 ] =

## nth order divided dierences

f [x0 , x1 , . . . , xn ] =
3.12.1

.
x0 xn

## 1. Divided dierence of a constant is zero

Let f (x) = c.
Then f [x0 , x1 ] =

f (x0 ) f (x1 )
cc
=
= 0.
x0 x1
x0 x1

## 2. Divided dierence of cf (x), c is constant, is the divided dierence of f (x) multiplied by c

Let g(x) = cf (x).
Therefore,
g[x0 , x1 ] =

g(x0 ) g(x1 )
cf (x0 ) cf (x1 )
f (x0 ) f (x1 )
=
=c
= cf [x0 , x1 ].
x0 x1
x0 x1
x0 x1

## 118 Numerical Analysis

3. Divided dierence is linear
Let h(x) = af (x) + bg(x).
Now,
h(x0 ) h(x1 )
af (x0 ) + bg(x0 ) af (x1 ) bg(x1 )
=
x0 x1
x0 x1
f (x0 ) f (x1 )
g(x0 ) g(x1 )
= a
+b
x0 x1
x0 x1
= af [x0 , x1 ] + bg[x0 , x1 ].

h[x0 , x1 ] =

## That is, divided dierence obeys linear property.

4. Divided dierences are symmetric functions
The rst order divided dierence is
f [x0 , x1 ] =

f (x0 ) f (x1 )
f (x1 ) f (x0 )
=
= f [x1 , x0 ].
x0 x1
x1 x0

## That is, rst order dierence is symmetric.

Also, f [x0 , x1 ] =

1
1
f (x0 ) +
f (x1 ).
x0 x1
x1 x0

## The second order dierence is

f [x0 , x1 ] f [x1 , x2 ]
x0 x2

1  1
1
=
f (x0 ) +
f (x1 )
x0 x2 x0 x1
x1 x0
 1

1

f (x1 ) +
f (x2 )
x1 x2
x2 x1
1
f (x0 )
=
(x0 x2 )(x0 x1 )
1
1
f (x1 ) +
f (x2 ).
+
(x1 x0 )(x1 x2 )
(x2 x0 )(x2 x1 )

f [x0 , x1 , x2 ] =

Interpolation

119

## Similarly, it can be shown that

1
f (x0 )
(x0 x1 )(x0 x2 ) (x0 xn )
1
f (x1 ) +
+
(x1 x0 )(x1 x2 ) (x1 xn )
1
f (xn )
+
(xn x0 )(xn x1 ) (xn xn1 )
n

f (xi )
.
=
(xi x0 ) (xi xi1 )(xi xi+1 ) (xi xn )

f [x0 , x1 , . . . , xn ] =

i=0

(3.64)
From these relations it is easy to observe that the divided dierences are symmetric.
5. For equispaced arguments, the divided dierences can be expressed in terms of
forward dierences.
That is,
f [x0 , x1 , . . . , xn ] =

1
hn .n!

n y0 .

## In this case, xi = x0 + ih, i = 0, 1, . . . , n.

f (x1 ) f (x0 )
y1 y 0
y0
.
=
=
x1 x0
x1 x0
h
f [x0 , x1 ] f [x1 , x2 ]
1  y0 y1 
f [x0 , x1 , x2 ] =

=
x0 x2
2h h
h
2
y0
y1 y0
=
.
=
2
2h
2!h2
f [x0 , x1 ] =

## The result is true for n = 1, 2. Let the result be true for

k y0
n = k, i.e., f [x0 , x1 , . . . , xk ] =
.
k! hk
Now,
f [x0 , x1 , . . . , xk ] f [x1 , x2 , . . . , xk+1 ]
x0 xk+1
 k y
1
k y1 
0
=

(xk+1 x0 ) k! hk
k! hk
1
[k y1 k y0 ]
=
(k + 1)k! hk+1
k+1 y0
=
.
(k + 1)! hk+1

f [x0 , x1 , . . . , xk , xk+1 ] =

## 120 Numerical Analysis

Hence, by mathematical induction
f [x0 , x1 , . . . , xn ] =

1
hn .n!

n y0 .

(3.65)

6. Divided dierences for equal arguments or divided dierences for conuent arguments.
If the arguments are equal then the divided dierences have a meaning. If two
arguments are equal then the divided dierence has no meaning as denominator
becomes zero. But, by limiting process one can dene the divided dierences for
equal arguments which is known as conuent divided dierences.
f [x0 , x0 ] = lim f [x0 , x0 + ] = lim
0

f (x0 + ) f (x0 )
= f  (x0 ),

## provided f (x) is dierentiable.

f [x0 , x0 , x0 ] = lim f [x0 , x0 , x0 + ] = lim
0

= lim

f  (x0 )

0
f (x0 +)f (x0 )

f [x0 , x0 ] f [x0 , x0 + ]


f  (x0 ) f (x0 + ) + f (x0 )  0
form
= lim
0
2
0


f (x0 ) f (x0 + )
= lim
(by LHospital rule)
0
2
f  (x0 )
=
.
2!
0



## Similarly, it can be shown that f [x0 , x0 , x0 , x0 ] =

f (x0 )
.
3!

In general,
(k+1) times




f k (x0 )
f [x0 , x0 , . . . , x0 ] =
.
k!
In other words,

(3.66)

(k+1) times




dk
f (x0 ) = k! f [x0 , x0 , . . . , x0 ].
k
dx

(3.67)

## 7. The nth order divided dierence of a polynomial of degree n is constant

Let f (x) = a0 xn + a1 xn1 + a2 xn2 + + an1 x + an , (a0 = 0) be a polynomial
of degree n.

Interpolation

121

Then
f (x) f (x0 )
x x0
xn xn0
xn1 x0n1
xn2 x0n2
x x0
= a0
+ a1
+ a2
+ + an1
x x0
x x0
x x0
x x0
= a0 [xn1 + xn2 x0 + xn3 x20 + + xx0n2 + x0n1 ]

f [x, x0 ] =

## +a1 [xn2 + xn3 x0 + xn4 x20 + + xxn3

+ xn2
] + + an1
0
0
= a0 xn1 + (a0 x0 + a1 )xn2 + (a0 x20 + a1 x0 + a2 )xn3 + + an1
= b0 xn1 + b1 xn2 + b2 xn3 + + bn1 ,
where b0 = a0 , b1 = a0 x0 + a1 , b2 = a0 x20 + a1 x0 + a2 , . . . , bn1 = an1 .
Thus, rst order divided dierence of a polynomial of degree n is a polynomial of
degree n 1.
Again, the second order divided dierence is
f [x0 , x] f [x0 , x1 ]
x x1
n1
x
xn1
xn2 xn2
xn3 x1n3
1
1
= b0
+ b1
+ b2
x x1
x x1
x x1
x x1
+ + bn2
x x1
n2
n2
= b0 [x
+x
x1 + xn4 x21 + + xx1n2 ]

f [x, x0 , x1 ] =

## +b1 [xn3 + xn4 x1 + xn5 x21

+ + xx1n4 + x1n3 ] + + bn2
= b0 xn2 + (b0 x1 + b1 )xn3 + (b0 x21 + b1 x1 + b2 )xn4
+ + bn2
= c0 xn2 + c1 xn3 + c2 xn4 + + cn2 ,
where c0 = b0 , c1 = b0 x1 + b1 , c2 = b0 x21 + b1 x1 + b2 , . . . , cn2 = bn2 .
This is a polynomial of degree n 2. So, the second order divided dierence is a
polynomial of degree n 2.
In this way, it can be shown that the n order divided dierence of a polynomial
of degree n is constant and which is equal to a0 .

3.13

## Suppose the function y = f (x) is known at the points x0 , x1 , . . . , xn and yi = f (xi ),

i = 0, 1, . . . , n. The points xi , i = 0, 1, . . . , n need not be equispaced.

## 122 Numerical Analysis

Now, from the denition of divided dierence of second order
f [x0 , x] =

f (x) f (x0 )
.
x x0

## Then f (x) = f (x0 ) + (x x0 )f [x0 , x].

From third order divided dierence,
f [x0 , x1 ] f [x0 , x]
x1 x
i.e., f [x0 , x] = f [x0 , x1 ] + (x x1 )f [x0 , x1 , x]
f (x) f (x0 )
= f [x0 , x1 ] + (x x1 )f [x0 , x1 , x]
i.e.,
x x0
Thus, f (x) = f (x0 ) + (x x0 )f [x0 , x1 ] + (x x0 )(x x1 )f [x0 , x1 , x].
f [x0 , x1 , x] =

## Similarly, for the arguments x0 , x1 , x2 , x,

f (x) = f (x0 ) + (x x0 )f [x0 , x1 ] + (x x0 )(x x1 )f [x0 , x1 , x2 ]
+ (x x0 )(x x1 )(x x2 )f [x0 , x1 , x2 , x3 ] +
+ (x x0 )(x x1 ) (x xn )f [x0 , x1 , x2 , . . . , xn , x].

(3.68)

This formula is known as Newtons fundamental or Newtons general interpolation formula including error term.
Tabular form of divided dierences are shown in Table 3.4.
Table 3.4: Divided dierence table.
x
x0
x0 x1
x0 x2

x1
x1 x2

x1 x3

x2
x2 x3
x3

..
.
..
.
..
.
..
.
..
.
..
.
..
.
..
.

f (x)

First

Second

Third

f (x0 )
f [x0 , x1 ]
f (x1 )

f [x0 , x1 , x2 ]
f [x1 , x2 ]

f (x2 )

f [x1 , x2 , x3 ]
f [x2 , x3 ]

f (x3 )

f [x0 , x1 , x2 , x3 ]

Interpolation

123

Error term
The error term
E(x) = (x x0 )(x x1 ) (x xn )f [x0 , x1 , . . . , xn , x]
f n+1 ()
, [using (3.66)]
= (x x0 )(x x1 ) (x xn )
(n + 1)!
where min{x0 , x1 , . . . , xn , x} < < max{x0 , x1 , . . . , xn , x}.
Example 3.13.1 Find the value of y when x = 1.5 from the following table:
x
y

:
:

1
0.6931

5
1.7918

7
2.0794

10
2.3979

12
2.5649

2nd

3rd

## using Newtons divided dierence formula.

Solution. The divided dierence table is
x
1

y
0.6931

1.7918

4
6
9
11

0.0218
0.1438

2.0794

3
5

4th

0.2747

1st

0.0016
0.0075

0.1062
10

2.3979

0.0001
0.0004

0.0045
0.0835

12

2.5649

## Here x = 1.5, x0 = 1, x1 = 5, x2 = 7, x3 = 10, x4 = 12,

f [x0 ] = 0.6931, f [x0 , x1 ] = 0.2747, f [x0 , x1 , x2 ] = 0.0218,
f [x0 , x1 , x2 , x3 ] = 0.0016, f [x0 , x1 , x2 , x3 , x4 ] = 0.0001.
The Newtons divided dierence formula is
y(1.5) = f [x0 ] + (x x0 )f [x0 , x1 ] + (x x0 )(x x1 )f [x0 , x1 , x2 ]
+(x x0 )(x x1 )(x x2 )f [x0 , x1 , x2 , x3 ]
+(x x0 )(x x1 )(x x2 )(x x3 )f [x0 , x1 , x2 , x3 , x4 ]
= 0.6931 + 0.1374 + 0.0382 + 0.0154 + 0.0082
= 0.8922.

(3.69)

3.14

## Deductions of other Interpolation Formulae from Newtons

Divided Dierence Formula

3.14.1

## If the arguments x0 , x1 , . . . , xn are equispaced, then xi = x0 + ih, i = 0, 1, . . . , n, and

n f (x0 )
. [see (3.65)]
n! hn
In this particular case, the Newtons divided dierence formula becomes
f [x0 , x1 , . . . , xn ] =

f (x0 )
2 f (x0 )
+ (x x0 )(x x1 )
1!h
2!h2
3
f (x0 )
+ (x x0 )(x x1 )(x x2 )
+
3!h3
n f (x0 )
+ E(x),
+ (x x0 )(x x1 ) (x xn1 )
n!hn

(x) = f (x0 ) + (x x0 )

where
E(x) = (x x0 )(x x1 ) (x xn )f [x, x0 , x1 , . . . , xn ]
n+1 f ()
= (x x0 )(x x1 ) (x xn )
.
(n + 1)! hn+1
Now, a unit less quantity u =
formula.
So, x xi = (u i)h.
Then

x x0
, i.e., x = x0 + uh is introduced to simplify the
h

u(u 1) 2
f (x0 ) +
2!
u(u 1)(u 2) (u n + 1) n
f (x0 ) + E(x),
+
n!
f n+1 ()
,
E(x) = u(u 1)(u 2) (u n)
(n + 1)!
min{x, x0 , x1 , . . . , xn } < < max{x, x0 , x1 , . . . , xn }.
(x) = f (x0 ) + uf (x0 ) +

3.14.2

## Newtons backward dierence interpolation formula

Let
(x) = f (xn ) + (x xn )f [xn , xn1 ] + (x xn )(x xn1 )f [xn , xn1 , xn2 ]
+ + (x xn )(x xn1 ) (x x1 )f [xn , xn1 , . . . , x1 , x0 ]
+E(x),

Interpolation

where
E(x) = (x xn )(x xn1 ) (x x1 )(x x0 )f [x, xn , xn1 , . . . , x1 , x0 ].
From the relation (3.65), we have
f [xn , xn1 , . . . , xnk ] =

k f (xnk )
k f (xn )
=
.
k!hk
k!hk

Therefore,
f (xn )
2 f (xn )
+ (x xn )(x xn1 )
+
1!h
2!h2
n f (xn )
+ E(x),
+ (x xn )(x xn1 ) (x x1 )(x x0 )
n!hn

(x) = f (xn ) + (x xn )

where
E(x) = (x xn )(x xn1 ) (x x1 )(x x0 )

n+1 f ()
,
(n + 1)!hn+1

3.14.3

## From the denition of (n + 1) order divided dierence (3.64) we have

f [x0 , x1 , . . . , xn ] =

n

i=0

f (xi )
.
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )

For (n + 2) arguments x, x0 , . . . , xn ,
f [x, x0 , x1 , . . . , xn ] =
+

f (x)
(x x0 )(x x1 ) (x xn )
n

f (xi )
i=0

f (x) 
f (xi )
+
,
w(x)
(xi x)w (xi )
n

i=0

## where w(x) = (x x0 )(x x1 ) (x xn ).

Therefore,
f (x) =

n

i=0

n

i=0

w(x)f (xi )
+ w(x)f [x, x0 , x1 , . . . , xn ]
(x xi )w (xi )
Li (x)f (xi ) + w(x)

f n+1 ()
[using (3.65)]
(n + 1)!

125

## 126 Numerical Analysis

where min{x0 , x1 , . . . , xn , x} < < max{x0 , x1 , . . . , xn , x} and
w(x)
Li (x) =
.
(x xi )w (xi )
This is the Lagranges interpolation formula with error term.
Note 3.14.1 It is observed that the divided dierences for equispaced arguments produce the Newton forward and backward dierence formulae. Also, this interpolation
gives Lagranges interpolation formula.
Actually, Lagranges interpolation formula and Newtons divided dierence interpolation formula are equivalent. This fact is proved in the following.

3.15

## Equivalence of Lagranges and Newtons divided

dierence formulae

In the following, it is proved that the Lagranges interpolation formula and Newtons
divided dierence interpolation formula are equivalent.
The Lagranges interpolation polynomial for the points (xi , yi ), i = 0, 1, . . . , n of
degree n is
(x) =

n


Li (x)yi ,

(3.70)

i=0

where
Li (x) =

## (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn )

.
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )

(3.71)

## The Newtons interpolation formula with divided dierence is given by

(x) = f (x0 ) + (x x0 )f [x0 , x1 ] + (x x0 )(x x1 )f [x0 , x1 , x2 ] +
+(x x0 )(x x1 ) (x xn1 )f [x0 , x1 , , xn ]

f (x0 )
f (x1 )
+
= f (x0 ) + (x x0 )
x0 x1 x1 x0
+(x x0 )(x x1 )

f (x1 )
f (x2 )
f (x0 )
+
+
(x0 x1 )(x0 x2 ) (x1 x0 )(x1 x2 ) (x2 x0 )(x2 x1 )
+ + (x x0 )(x x1 ) (x xn1 )

f (xn )
f (x0 )
+ +
.
(3.72)
(x0 x1 ) (x0 xn )
(xn x0 ) (xn xn1 )

Interpolation

127

## The coecient of f (x0 ) in the above expression is

(x x0 )(x x1 )
x x0
+
+
x0 x1 (x0 x1 )(x0 x2 )
(x x0 )(x x1 ) (x xn1 )
+
(x0 x1 )(x0 x2 ) (x0 xn )

(x x0 )(x x2 )
x x0
(x x1 )
+
1+
+
(x0 x1 )
x0 x2 (x0 x2 )(x0 x3 )

(x x0 )(x x2 ) (x xn1 )
+ +
(x0 x2 )(x0 x3 ) (x0 xn )

(x x0 )(x x3 ) (x xn1 )
(x x1 )(x x2 )
x x0
+ +
1+
(x0 x1 )(x0 x2 )
x0 x2
(x0 x3 )(x0 x4 ) (x0 xn )

x x0
(x x1 )(x x2 ) (x xn1 )
1+
(x0 x1 )(x0 x2 ) (x0 xn1 )
x0 xn
(x x1 )(x x2 ) (x xn1 )(x xn )
(x0 x1 )(x0 x2 ) (x0 xn1 )(x0 xn )
L0 (x).
1+

=
=
=
=

Similarly, it can be shown that the coecient of f (x1 ) is L1 (x), coecient of f (x2 )
is L2 (x) and so on.
Thus, (3.72) becomes
(x) = L0 (x)f (x0 ) + L1 (x)f (x1 ) + + Ln (x)f (xn ) =

n


Li (x)f (xi ).

i=1

Thus the Lagranges interpolation and Newtons divided dierence interpolation formulae are equivalent.
Example 3.15.1 A function y = f (x) is given at the points x = x0 , x1 , x2 . Show
that the Newtons divided dierence interpolation formula and the corresponding
Lagranges interpolation formula are identical.
Solution. The Newtons divided dierence formulae is given as
y = f (x) = y0 + (x x0 )f [x0 , x1 ] + (x x0 )(x x1 )f [x0 , x1 , x2 ]
f (x1 ) f (x0 )
+ (x x0 )(x x1 )
= y0 + (x x0 )
x1 x0

f (x0 )
f (x1 )
f (x2 )
+
+

## 128 Numerical Analysis

(x x0 )(x x1 )
(x0 x)
+
f (x0 )
= 1
(x0 x1 ) (x0 x1 )(x0 x2 )

(x x0 )(x x1 )
(x x0 )
(x x0 )(x x1 )
+
f (x1 ) +
f (x2 )
+
(x1 x0 ) (x1 x0 )(x1 x2 )
(x2 x0 )(x2 x1 )
(x x0 )(x x2 )
(x x1 )(x x2 )
f (x0 ) +
f (x1 )
=
(x0 x1 )(x0 x2 )
(x1 x0 )(x1 x2 )
(x x0 )(x x1 )
f (x2 )
+
(x2 x0 )(x2 x1 )

## which is the Lagranges interpolation polynomial. Hence Newtons divided dierence

interpolation polynomial and Lagranges interpolation polynomial are equivalent.
Example 3.15.2 For the following table, nd the interpolation polynomial using
(i) Lagranges formula and (ii) Newtons divided dierence formula, and hence show
that both represent same interpolating polynomial.
x
f (x)

:
:

0
3

2
8

4
11

8
18

(x 0)(x 4)(x 8)
(x 2)(x 4)(x 8)
3+
8
(0 2)(0 4)(0 8)
(2 0)(2 4)(2 8)
(x 0)(x 2)(x 4)
(x 0)(x 2)(x 8)
11 +
19
+
(4 0)(4 2)(4 8)
(8 0)(8 2)(8 4)
x3 12x2 + 32x
x3 14x2 + 56x 64
3+
8
=
64
24
x3 10x2 + 16x
x3 6x2 + 8x
+
11 +
19
32
192
1 3 1 2 10
=
x x + x + 3.
24
2
3

(x) =

x

f (x)

0
2
4
8

3
8
11
19

1st divided
dierence

2nd divided
dierence

3rd divided
dierence

5/2
3/2
2

1/4
1/12

1/24

Interpolation

129

## Newtons divided dierence polynomial is

1
5
+ (x 0)(x 2) ( )
2
4
1
+(x 0)(x 2)(x 4)
24
1 3
5
1 2
= 3 + x (x 2x) + (x 6x2 + 8x)
2
4
24
1 3 1 2 10
x x + x + 3.
=
24
2
3

(x) = 3 + (x 0)

Thus, it is observed that the interpolating polynomial by both Lagranges and Newtons divided dierence formulae are one and same.
It may be noted that Newtons formula involves less number of arithmetic operations
than that of Lagranges formula.
Algorithm 3.4 (Divided dierence). This algorithm nds the value of y at a
given point x from a table (xi , yi ), i = 0, 1, 2, . . . , n, by Newtons divided dierence
interpolation formula.
Let

yk = f (xk ) = dk,0 .
dk,0 dk1,0
f (xk ) f (xk1 )
=
f [xk , xk1 ] = dk,1 =
xk xk1
xk xk1
dk,1 dk1,1
f [xk , xk1 ] f [xk1 , xk2 ]
f [xk , xk1 , xk2 ] = dk,2 =
=
xk xk2
xk xk2
dk,2 dk1,2
f [xk , xk1 , xk2 , xk3 ] = dk,3 =
xk xk3

In general,
dk,i =

dk,i1 dk1,i1
, i = 1, 2, . . . , n.
xk xki

(3.73)

Using the above notations, the Newtons divided dierence formula (3.68) can be
written in the following form.
f (x) = d0,0 + d1,1 (x x0 ) + d2,2 (x x0 )(x x1 )
+d3,3 (x x0 )(x x1 )(x x2 ) +
+dn,n (x x0 )(x x1 ) (x xn1 )

## = dn,n (x xn1 )(x xn2 )(x xn3 ) (x x0 )

+dn1,n1 (x xn2 )(x xn3 ) (x x0 )
+dn2,n2 (x xn3 ) (x x0 ) + d1,1 (x x0 ) + d0,0
= ( (((dn,n (x xn1 ) + dn1,n1 )(x xn2 )
+dn2,n2 )(x xn3 ) + dn3,n3 )(x xn4 ) + ) + d0,0
Algorithm Divided dierence
Step 1: Read n; //the degree of the polynomial//
Read (x(i), y(i)), i = 0, 1, . . . , n //the points//
Step 2: for k = 0 to n do
Set d(k, 0) = y(k);
Step 3: for i = 0 to n do
//compute all divided dierences//
for k = i to n do
d(k, i) = (d(k, i 1) d(k 1, i 1))/(x(k) x(k i));
//for which the polynomial is to be calculated//
Step 5: Set sum = d(n, n)
//initialization of sum//
Step 6: for k = n 1 to 0 do
Compute sum = sum (xg x(k)) + d(k, k);
Step 7: Print xg, sum;
end Divided dierence
Program 3.4
.
/* Program Divided Difference Interpolation
This program finds the value of y=f(x) at a given x when
the function is supplied as (x[i],y[i]), i=0, 1, ..., n.
*/
#include<stdio.h>
#include<math.h>
void main()
{
int i,k,n;
float x,y,xg,sum,d;
printf("Enter number of subintervals ");
scanf("%d",&n);
printf("Enter x and y values ");
for(i=0;i<=n;i++)
scanf("%f %f",&x[i],&y[i]);
printf("Enter interpolating point x ");
scanf("%f",&xg);
printf("The given values of x and y are\nx-value y-value\n");

Interpolation

131

for(i=0;i<=n;i++) printf("%f
%f\n",x[i],y[i]);
for(k=0;k<=n;k++) d[k]=y[k];
for(i=1;i<=n;i++)
for(k=i;k<=n;k++)
d[k][i]=(d[k][i-1]-d[k-1][i-1])/(x[k]-x[k-i]);
sum=d[n][n];
for(k=n-1;k>=0;k--) sum=sum*(xg-x[k])+d[k][k];
printf("The interpolated value at x=%f is %f ",xg,sum);
}
A sample of input/output:
Enter number of subintervals 4
Enter x and y values
0.10
1.1052
0.15
1.1618
0.20
1.2214
0.25
1.2840
0.30
1.3499
Enter interpolating point x 0.12
The given values of x and y are
x-value
y-value
0.100000
1.105200
0.150000
1.161800
0.200000
1.221400
0.250000
1.284000
0.300000
1.349900
The interpolated value at x=0.120000 is 1.127468

3.16

Inverse Interpolation

In interpolation, for a given set of values of x and y, the value of y is determined for a
given value of x. But the inverse interpolation is the process which nds the value of
x for a given y. Commonly used inverse interpolation formulae are based on successive
iteration.
In the following, three inverse interpolation formulae based on Lagrange, Newton
forward and Newton backward interpolation formulae are described. The inverse interpolation based on Lagranges formula is a direct method while the formulae based on
Newtons interpolation formulae are iterative.

3.16.1

y=

n

i=0

w(x) yi
.
(x xi )w (xi )

## When x and y are interchanged then the above relation changes to

x=

n

i=0


w(y)xi
=
Li (y)xi ,
(y yi )w (yi )
n

i=0

where
Li (y) =

w(y)
(yy0 )(yy1 ) (yyi1 )(yyi+1 ) (yyn )
=
.
(yyi )w (yi ) (yi y0 )(yi y1 ) (yi yi1 )(yi yi+1 ) (yi yn )

This formula gives the value of x for given value of y and the formula is known as
Lagranges inverse interpolation formula.
3.16.2

## Based on Newtons forward dierence interpolation formula

The Newtons forward dierence interpolation formula is
u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0 +
2!
3!
u(u 1)(u 2) (u n 1) n
+
y0 ,
n!

y = y0 + uy0 +

x x0
.
where u =
h
The above formula can be written as
u(u 1) 2
u(u 1)(u 2) 3
1 
y0
y0
u =
y y0
y0
2!
3!
u(u 1)(u 2) (u n 1) n 
y0 .

n!
Let the rst approximation of u be denoted by u(1) and it is obtained by neglecting
the second and higher dierences as
u(1) =

1
(y y0 ).
y0

Interpolation

133

Next, the second approximation, u(2) , is obtained by neglecting third and higher order
dierences as follows:
u(2) =

1 
u(1) (u(1) 1) 2 
y y0
y0 .
y0
2!

## Similarly, the third approximation u(3) is given by

u(3) =

1 
u(2) (u(2) 1) 2
u(2) (u(2) 1)(u(2) 2) 3 
y0
y0 .
y y0
y0
2!
3!

In general,
u

(k+1)

1
u(k) (u(k) 1) 2
u(k) (u(k) 1)(u(k) 2) 3
y y0
y0
y0
=
y0
2!
3!

## u(k) (u(k) 1) (u(k) k) k+1

y0 ,

(k + 1)!

k = 0, 1, 2, . . . .
This process of approximation should be continued till two successive approximations
(k+1)
and u(k) be equal up to desired number of decimal places. Then the value of x is
u
obtained from the relation x = x0 + u(k+1) h.
Example 3.16.1 From the table of values
x
y

:
:

1.8
3.9422

2.0
4.6269

2.2
5.4571

2.4
6.4662

2.6
7.6947

## nd x when y = 5.0 using the method of successive approximations.

Solution. The dierence table is
x
1.8

y
3.9422

2.0

4.6269

2 y

3 y

0.6847
0.1455
0.8302
2.2

5.4571

2.4

6.4662

2.6

7.6947

0.0334
0.1789

1.0091

0.0405
0.2194

1.2285

## 134 Numerical Analysis

Let x0 = 2.0, h = 0.2. The value of u is determined by successive approximation.
The rst approximation is
u(1) =
u(2) =
=
u(3) =
=

1
1
(5.0 4.6269) = 0.4494.
(y y0 ) =
y0
0.8302

1
u(1) (u(1) 1) 2
u(1) (u(1) 1) 2 y0
y0 = u(1)
y y0
y0
2!
2!
y0
0.4494(0.4494 1) 0.1789
= 0.4761.
0.4494
2
0.8302
u(2) (u(2) 1) 2 y0 u(2) (u(2) 1)(u(2) 2) 3 y0
u(1)

2
y0
3!
y0
0.4761(0.4761 1) 0.1789

0.4494
2
0.8302
0.4761(0.4761 1)(0.4761 2) 0.0405

6
0.8302
0.4494 + 0.0269 0.0031 = 0.4732.

3.16.3

## The Newtons backward formula is

v(v + 1) 2
v(v + 1)(v + 2) 3
yn +
yn +
2!
3!
v(v + 1)(v + 2) (v + n 1) n
yn ,
+
n!

y = yn + vyn +

where v =
That is,

x xn
or x = xn + vh.
h
v =

1 
v(v + 1) 2
v(v + 1)(v + 2) 3
yn
yn
y yn
yn
2!
3!
v(v + 1) (v + n 1) n 
yn .

n!

Neglecting second and higher order dierences, the rst approximation is given by
v (1) =
Similarly,
v (2) =

1
(y yn ).
yn

1 
v (1) (v (1) + 1) 2 
y yn
yn .
yn
2!

Interpolation

v (3) =

135

1 
v (2) (v (2) + 1) 2
v (2) (v (2) + 1)(v (2) + 2) 3 
y yn
yn
yn
yn
2!
3!

and so on.
In general,
v

(k+1)

1
v (k) (v (k) + 1) 2
v (k) (v (k) + 1)(v (k) + 2) 3
yn
yn
y yn
=
yn
2!
3!

## v (k) (v (k) + 1) (v (k) + k) k+1

yn ,

(k + 1)!

k = 0, 1, 2, . . . .
This iteration continues until two consecutive values v (k) and v (k+1) become equal up
to a desired number of signicant gures.
The value of x is given by x = xn + v (k+1) h.

3.16.4

## Use of inverse interpolation to nd a root of an equation

Suppose x = be a root of the equation f (x) = 0 and let it lies between a and b, i.e.,
a < < b. Now, a table is constructed for some values of x, within (a, b), and the
corresponding values of y. Then by inverse interpolation, the value of x is determined
when y = 0. This value of x is the required root.
Example 3.16.2 Find a real root of the equation x3 3x + 1 = 0.
Solution. Let y = x3 3x + 1. One root of this equation lies between 1/4 and 1/2.
Let us consider the points x = 0.25, 0.30, 0.35, 0.40, 0.45, 0.50. The table is shown
below.
x
0.25
0.30
0.35
0.40
0.45
0.50

y
0.265625
0.127000
0.007125
0.136000
0.258875
0.375000

2 y

3 y

0.138625
0.134125
0.128875
0.122875
0.116125

0.00450
0.00525
0.00600
0.00675

0.00075
0.00075
0.00075

## 136 Numerical Analysis

Now,
y0
0.265625
= 1.916140.
=
y0
0.138625

1
u(1) (u(1) 1) 2
y0 +
y0

y0
2

1
1.916140 0.916140
0.265625 +
0.00450
0.138625
2
1.944633.

1
u(2) (u(2) 1) 2
u(2) (u(2) 1)(u(2) 2) 3
y0 +
y0 +
y0

y0
2!
3!

1
1.944633 0.944633
0.265625 +
0.00450
0.138625
2

+
0.000750
6
1.945864.

u(1) =
u(2) =
=
=
u(3) =
=

3.17

## Choice and use of Interpolation Formulae

If the interpolating points are not equally spaced then Lagranges, Newtons divided
dierence or Aitkens iterated interpolation formulae may be used. Newtons forward
formula is appropriate for interpolation at the beginning of the table, Newtons backward formula for interpolation at the end of the table, Stirlings or Bessels formula for
interpolation at the centre of the table. It is well known that the interpolation polynomial is unique and the above formulae are just dierent forms of one and the same
interpolation polynomial and the results obtained by the dierent formulae should be
identical. Practically, only a subset of the set of given interpolating points in the table is used. For interpolation at the beginning of the table, it is better to take this
subset from the beginning of the table. This reason recommends the use of Newtons
forward formula for interpolation at the beginning of the table. For interpolation, near
the end of the table, interpolating points should be available at the end of the table and
hence Newtons backward formula is used for interpolation at the end of the table. For
the same reasons the central dierence formulae like Stirlings, Bessels, Everetts etc.
are used for interpolation near the centre of the table. The proper choice of a central
interpolation formulae depends on the error terms of the dierent formulae.
For interpolation near the centre of the table, Stirlings formula gives the most accurate result for 1/4 s 1/4, and Bessels formula gives most accurate result near

Interpolation

137

s = 1/2, i.e., for 1/4 s 3/4. If all the terms of the formulae are considered, then
both the formulae give identical result. But, if some terms are discarded to evaluate the
polynomial, then Stirlings and Bessels formulae, in general, do not give the same result
and then a choice must be made between them. The choice depends on the order of the
highest dierence that could be neglected so that contributions from it and further differences would be less than half a unit in the last decimal place. If the highest dierence
is of odd order, then Stirlings formula is used and if it is of even order, then, generally,
Bessels formula is used. This conclusion is drawn from the following comparison.
The term of Stirlings formula containing the third dierences is
s(s2 12 ) 3 y1 + 3 y2
.
6
2
This term may be neglected if its magnitude is less than half a unit in the last place,
i.e., if

 2
 s(s 12 ) 3 y1 + 3 y2  1
 .

 2

6
2
 2

 s(s 1) 


 is 0.064 at s = 1/ 3. Then
The maximum value of 

6


 3

3
3
3




0.064 y1 + y2  < 1 , i.e.,  y1 + y2  < 7.8.




2
2
2
The term containing third order dierence of Bessels formula will be less than half
a unit in the last place if


 1
 s(s 1)(s 1/2) 3

y1  < .

6
2


 s(s 1)(s 1/2) 

 is 0.008 and so that |3 y1 | < 62.5.
The maximum value of 

6
Thus, if the third dierence is ignored, Bessels formula gives about eight times more
accurate result than Stirlings formula. But, if the third dierences need to be retained
and when the magnitude is more than 62.5, then Everetts formula is more appropriate.
It may be reminded that the Bessels formula with third dierences is equivalent to
Everetts formula with second dierences.
Depending on these discussions the following working rules are recommended for use
of interpolation formulae.
(i) If the interpolating point is at the beginning of the table, then use Newtons
forward formula with a suitable starting point x0 such that 0 < u < 1.

## 138 Numerical Analysis

(ii) If the interpolating point is at the end of the table, then use Newtons backward
formula with a suitable starting point xn such that 1 < u < 0.
(iii) If the interpolating point is at the centre of the table and the dierence table ends
with odd order dierences, then use Stirlings formula.
(iv) If the interpolating point is at the centre of the table and the dierence table ends
with even order dierences, then use Bessels or Everetts formula.
(v) If the arguments are not equispaced then use Lagranges formula or Newtons
divided dierence formula or Aitkens iterative formula.
In some functions, a higher degree interpolation polynomial does not always give
the best result compared to a lower degree polynomial. This fact is illustrated in the
following example.
1
. In [3, 3] the second degree polynomial y = 2 (x) = 10.1x2 and
Let f (x) =
1 + x2
fourth degree polynomial y = 4 (x) = 0.15577u4 1.24616u3 +2.8904u3 1.59232u+0.1
where u = (x + 3)/1.5 are derived. The graph of the curves y = f (x), y = 2 (x) and
y = 4 (x) are shown in Figure 3.4.
y
6

y=4 (x)
=

y=2 (x)

-1.5

0 1.5

y=f (x)
- x

Figure 3.4: The graph of the curves y = f (x), y = 2 (x) and y = 4 (x).
For this function y = f (x), the fourth degree polynomial gives an absurd result at
x = 2. At this point f (2) = 0.2, 2 (2) = 0.6 and 4 (2) = 0.01539. It may be
noted that the functions y = f (x) and y = 2 (x) are positive for all values of x, but
y = 4 (x) is negative for some values of x. This example indicates that the higher
degree polynomial does always not give more accurate result.

Interpolation

3.18

139

## Hermites Interpolation Formula

The interpolation formulae considered so far make use of the function values at some
number of points, say, n + 1 number of points and an nth degree polynomial is obtained.
But, if the values of the function y = f (x) and its rst derivatives are known at n + 1
points then it is possible to determine an interpolating polynomial (x) of degree (2n+1)
which satises the (2n + 2) conditions

(xi ) = f (xi )
(3.74)
 (xi ) = f  (xi ), i = 0, 1, 2, . . . , n.
This formula is known as Hermites interpolation formula. Here, the number of
conditions is (2n + 2), the number of coecients to be determined is (2n + 2) and the
degree of the polynomial is (2n + 1).
Let us assume the Hermites interpolating polynomial in the form
(x) =

n


hi (x)f (xi ) +

i=0

n


Hi (x)f  (xi ),

(3.75)

i=0

where hi (x) and Hi (x), i = 0, 1, 2, . . . , n, are polynomial in x of degree at most (2n + 1).
Using conditions (3.74), we get

1, if i = j
(3.76)
hi (xj ) =
; Hi (xj ) = 0, for all i
0, if i = j
hi (xj ) = 0, for all i; Hi (xj ) =

1, if i = j
0, if i = j

(3.77)

Li (x) =

## (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn )

,
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )
i = 0, 1, 2, . . . , n.

Obviously,
Li (xj ) =

1, if i = j
0, if i = j.

(3.78)
(3.79)

Since each Li (x) is a polynomial of degree n, [Li (x)]2 is a polynomial of degree 2n.
Again, each Li (x) satises (3.79) and [Li (x)]2 also satises (3.79). Since hi (x) and Hi (x)
are polynomials in x of degree (2n + 1), their explicit form may be taken as
hi (x) = (ai x + bi )[Li (x)]2
Hi (x) = (ci x + di )[Li (x)]2

(3.80)

## 140 Numerical Analysis

Using the conditions (3.77), we obtain
ai xi + bi
ci xi + di
ai + 2Li (xi )
ci

=1
=0
=0
=1

(3.81)

## These relations give

bi = 1 + 2xi Li (xi )
ai = 2Li (xi ),
and
di = xi
ci = 1

(3.82)

## hi (x) = [2xLi (xi ) + 1 + 2xi Li (xi )][Li (xi )]2

= [1 2(x xi )Li (xi )][Li (x)]2 and
Hi (x) = (x xi )[Li (x)]2 .

(3.83)

Hence,

(x) =

n


i=0

n


(3.84)

i=0

## which is the required Hermites interpolation formula.

Example 3.18.1 Determine the Hermites polynomial ofdegree 5 which satises
the following data and hence nd an approximate value of 3 2.8.
x
y= 3x

y = 1/(3x2/3 )

:
:
:

1.5
1.14471
0.25438

2.0
1.25992
0.20999

2.5
1.35721
0.18096

Solution.
(x 2.0)(x 2.5)
(x x1 )(x x2 )
=
= 2x2 9x + 10,
(x0 x1 )(x0 x2 )
(1.5 2.0)(1.5 2.5)
(x x0 )(x x2 )
(x 1.5)(x 2.5)
=
= 4x2 + 16x 15,
L1 (x) =
(x1 x0 )(x1 x2 )
(2.0 1.5)(2.0 2.5)
(x 1.5)(x 2.0)
(x x0 )(x x1 )
=
= 2x2 7x + 6.
L2 (x) =
(x2 x0 )(x2 x1 )
(2.5 1.5)(2.5 2.0)
L0 (x) =

Interpolation

141

Therefore
L0 (x) = 4x 9, L1 (x) = 8x + 16, L2 (x) = 4x 7.
Hence L0 (x0 ) = 3, L1 (x1 ) = 0, L2 (x2 ) = 3.
h0 (x) =
=
h1 (x) =
h2 (x) =
=
H0 (x) =
H1 (x) =
H2 (x) =

## [1 2(x x0 )L0 (x0 )][L0 (x)]2 = [1 2(x 1.5)(3)][L0 (x)]2

(6x 8)(2x2 9x + 10)2 ,
[1 2(x x1 )L1 (x1 )][L1 (x)]2 = (4x2 16x + 15)2 ,
[1 2(x x2 )L2 (x2 )][L2 (x)]2 = [1 2(x 2.5)(3)](2x2 7x + 6)2
(16 6x)(2x2 7x + 6)2 .
(x x0 )[L0 (x)]2 = (x 1.5)(2x2 9x + 10)2 ,
(x x1 )[L1 (x)]2 = (x 2.0)(4x2 16x + 15)2 ,
(x x2 )[L2 (x)]2 = (x 2.5)(2x2 7x + 6)2 .

## The required Hermite polynomial is

(x) = (6x 8)(2x2 9x + 10)2 (1.14471) + (4x2 16x + 15)2 (1.25992)
+(16 6x)(2x2 7x + 6)2 (1.35721)
+(x 1.5)(2x2 9x + 10)2 (0.25438)
+(x 2)(4x2 16x + 15)2 (0.20999)
+(x 2.5)(2x2 7x + 6)2 (0.18096).

## To nd the value of 3 2.8, substituting x = 2.8 to the above relation.

Therefore,

3
2.8 = 10.07345 0.23040 + 1.25992 2.43360 1.08577 4.32640
+0.33069 0.23040 + 0.16799 2.43360 + 0.05429 4.32640
= 1.40948.

3.19

Spline Interpolation

Spline interpolation is very powerful and widely used method and has many applications
in numerical dierentiation, integration, solution of boundary value problems, two and
three - dimensional graph plotting etc. Spline interpolation method, interpolates a
function between a given set of points by means of piecewise smooth polynomials. In
this interpolation, the curve passes through the given set of points and also its slope
and its curvature are continuous at each point. The splines with dierent degree are
found in literature, among them cubic splines are widely used.

## 142 Numerical Analysis

3.19.1

Cubic spline

A function y(x) is called cubic spline in [x0 , xn ] if there exist cubic polynomials
p0 (x), p1 (x), . . . , pn1 (x) such that
y(x) = pi (x) on [xi , xi+1 ], i = 0, 1, 2, . . . , n 1.
pi1 (xi )
pi1 (xi )

=
=

## pi (xi ), i = 1, 2, . . . , n 1 (equal slope).

pi (xi ), i = 1, 2, . . . , n 1 (equal curvature).

and pi (xi ) = yi ,

pi (xi+1 ) = yi+1 , i = 0, 1, . . . , n 1.

(3.85)
(3.86)
(3.87)
(3.88)

It may be noted that, at the endpoints x0 and xn , no continuity on slope and curvature
are assigned. The conditions at these points are assigned, generally, depending on the
applications.
Let the interval [xi , xi+1 ], i = 0, 1, . . . , n 1 be denoted by ith interval.
Let hi = xi xi1 , i = 1, 2, . . . , n and Mi = y  (xi ), i = 0, 1, 2, . . . , n.
Let the cubic spline for the ith interval be
y(x) = ai (x xi )3 + bi (x xi )2 + ci (x xi ) + di ,

in [xi , xi+1 ].

(3.89)

## Since it passes through the end points xi and xi+1 , therefore,

yi = d i

(3.90)

and
yi+1 = ai (xi+1 xi )3 + bi (xi+1 xi )2 + ci (xi+1 xi ) + di
= ai h3i+1 + bi h2i+1 + ci hi+1 + di .

(3.91)

## Equation (3.89) is dierentiated twice and obtained the following equations.

y  (x) = 3ai (x xi )2 + 2bi (x xi ) + ci .


## and y (x) = 6ai (x xi ) + 2bi .

(3.92)
(3.93)


= 6ai hi+1 + 2bi ,
From (3.93), yi = 2bi and yi+1
that is, Mi = 2bi , Mi+1 = 6ai hi+1 + 2bi .
Therefore,

Mi
,
2
Mi+1 Mi
.
ai =
6hi+1
bi =

(3.94)
(3.95)

Interpolation

143

## Using (3.90), (3.94) and (3.95), equation (3.91) becomes

Mi+1 Mi 3
Mi 2
h
hi+1 +
+ ci hi+1 + yi
6hi+1
2 i+1
yi+1 yi 2hi+1 Mi + hi+1 Mi+1
.
i.e., ci =

hi+1
6
yi+1 =

(3.96)

## Thus, the coecients ai , bi , ci and di of (3.89) are determined in terms of n + 1

unknowns M0 , M1 , . . . , Mn . These unknowns are determined in the following way.
The condition of equation (3.86) state that the slopes of the two cubics pi+1 and pi
are same at xi .
Now, for the ith interval
yi = 3ai (xi xi )2 + 2bi (xi xi ) + ci = ci ,

(3.97)

## and for the (i 1)th interval

yi = 3ai1 (xi xi1 )2 + 2bi1 (xi xi1 ) + ci1 .

(3.98)

## Equating (3.97) and (3.98), we obtain

ci = 3ai1 h2i + 2bi1 hi + ci1 .
The values of ai1 , bi1 , ci1 and ci are substituted to the above equation and obtained
the following equation.
2hi+1 Mi + hi+1 Mi+1
yi+1 yi

hi+1
6


yi yi1 2hi Mi1 + hi Mi
Mi Mi1 2  Mi 
.
hi +

= 3
hi +
6hi
2
hi
6
After simplication the above equation reduces to
y
yi yi1 
i+1 yi

.
hi+1
hi

(3.99)

## This relation is true for i = 1, 2, . . . , n 1. Thus n 1 equations are available for

the n + 1 unknown quantities M0 , M1 , . . . , Mn . Now, two more conditions are required
to solve these equations uniquely. These conditions can be assumed to take one of the
following forms:
(i) M0 = Mn = 0. If this conditions are satised then the corresponding spline is
called natural spline.

## 144 Numerical Analysis

(ii) M0 = Mn , M1 = Mn+1 , y0 = yn , y1 = yn+1 , h1 = hn+1 . The corresponding spline
satisfying these conditions is called periodic spline.
(iii) y  (x0 ) = y0 , y  (xn ) = yn , i.e.,

6  y1 y 0
y0
h1
h1
6  
yn yn1 
yn
.
=
hn
hn

2M0 + M1 =
and Mn1 + 2Mn

(3.100)

## The corresponding spline satisfying the above conditions is called non-periodic

spline or clamped cubic spline.
h1 (M2 M1 )
hn (Mn1 Mn2 )
and Mn = Mn1 +
. The correh2
hn1
sponding spline is called extrapolated spline.

(iv) If M0 = M1

(v) If M0 = y0 and Mn = yn are specied. If a spline satisfy these conditions then it
y
yi yi1 
i+1 yi
Let Ai = hi , Bi = 2(hi + hi+1 ), Ci = hi+1 and Di = 6
.

hi+1
hi
Then (3.99) becomes
Ai Mi1 + Bi Mi + Ci Mi+1 = Di , i = 1, 2, . . . , n 1.

(3.101)

## The system of equations (3.101) is basically a tri-diagonal system. Depending on

dierent conditions, the tri-diagonal systems are dierent and they are stated below.
For natural spline, the tri-diagonal system for M1 , M2 , . . . , Mn1 is

0
0
B1 C1 0 0 0
M1
D1
A2 B2 C2 0 0

0
0 M D

2
2

0 A3 B3 C3 0

0
0
(3.102)
.. = ..

. .

Mn1
Dn1
0 0 0 0 0 An1 Bn1
and M0 = Mn = 0.
Imposing the conditions for non-periodic spline, we nd
2M0 + M1 = D0
and Mn1 + 2Mn = Dn ,


6 y1 y 0
where D0 =
y0
h1
h1


6
yn yn1

and Dn =
y
.
hn n
hn

(3.103)
(3.104)

(3.105)

Interpolation

145

Then equations (3.101), (3.103), (3.104) and (3.105) result the following tri-diagonal
system for the unknowns M0 , M1 , . . . , Mn .
M D

0
0
2 1 0 0 0
0
0
D1
M

A1 B1 C1 0 0
1
0
0

D2
M2

0 A2 B2 C2 0
0
0

. = .

.
(3.106)
. .

. .

## 0 0 0 0 An1 Bn1 Cn1

Mn1 Dn1
0 0 0 0 0
1
2
Mn
Dn
For the extrapolated spline the values of M0 and Mn are given by the relations
M0 = M1

h1 (M2 M1 )
hn (Mn1 Mn2 )
and Mn = Mn1 +
.
h2
hn1

(3.107)

## The rst expression is rewritten as

A1 h1
A1 h1
+ M2 C1
= D1 or, M1 B1 + M2 C1 = D1
M1 A1 + B1 +
h2
h2
A1 h1
A1 h1
and C1 = C1
.
h2
h2

= Dn1
Similarly, the second expression is transferred to Mn2 An1 + Mn1 Bn1
where
Cn1 hn
hn Cn1

An1 = An1
and Bn1
= Bn1 + Cn1 +
.
hn1
hn1
For this case, the tri-diagonal system of equations for M1 , M2 , . . ., Mn1 is


0
0
B1 C1 0 0 0
M1
D1
A2 B2 C2 0 0
0
0

M2

D2
0 A3 B3 C3 0

0
0
=
.
(3.108)

.
.

. .
.
.

Mn1
Dn1

0 0 0 0 0 An1 Bn1
where B1 = A1 + B1 +

## The values of M0 and Mn are obtained from the equation (3.107).

For the endpoint curvature-adjusted spline, the values of M0 and Mn are respectively
y0 and yn , where y0 and yn are specied. The values of M1 , M2 , . . ., Mn1 are given
by solving the following tri-diagonal system of equations

D1
0
0
B1 C1 0 0 0
M1
D2

A2 B2 C2 0 0
0
0

M2

.. ,
. =
0 A3 B3 C3 0
0
0
(3.109)

. .

.

Dn2
Mn1
0 0 0 0 0 An1 Bn1
D
n1

where

D1

= D1

A1 y0 ,


Dn1

= Dn1

Cn1 yn .

## 146 Numerical Analysis

Example 3.19.1 Fit a cubic spline curve that passes through (0, 0.0), (1, 0.5), (2,
2.0) and (3, 1.5) with the natural end boundary conditions, y  (0) = y  (3) = 0.
Solution. Here the intervals are (0, 1) (1, 2) and (2, 3), i.e., three intervals of
x, in each of which we can construct a cubic spline. These piecewise cubic spline
polynomials together gives the cubic spline curve y(x) in the entire interval (0, 3).
Here h1 = h2 = h3 = 1.
Then equation (3.99) becomes
Mi1 + 4Mi + Mi+1 = 6(yi+1 2yi + yi1 ),

i = 1, 2, 3.

That is,
M0 + 4M1 + M2 = 6(y2 2y1 + y0 ) = 6 (2.0 2 0.5 + 0.0) = 6
M1 + 4M2 + M3 = 6(y3 2y2 + y1 ) = 6 (1.5 2 2.0 + 0.5) = 12.
Imposing the conditions M0 = y  (0) = 0 and M3 = y  (3) = 0 to the above equations,
and they simplify as
4M1 + M2 = 6,

M1 + 4M2 = 12.

12
18
and M2 = .
These equations give M1 =
5
5
Let the natural cubic spline be given by
pi (x) = ai (x xi )3 + bi (x xi )2 + ci (x xi ) + di
where the coecients ai , bi , ci and di are given by the relations
Mi+1 Mi
Mi
,
,
bi =
6hi+1
2
yi+1 yi 2hi+1 Mi + hi+1 Mi+1
,

ci =
hi+1
6

ai =

d i = yi ,

for i = 0, 1, 2.
Therefore,
M1 M 0
= 0.4,
6
y1 y0 2M0 + M1

= 0.1,
c0 =
1
6
M2 M 1
= 1,
a1 =
6
a0 =

b0 =

M0
= 0,
2

d0 = y0 = 0.
b1 =

6
M1
= ,
2
5

Interpolation

y2 y1 2M1 + M2

= 1.3,
1
6
3
M3 M 2
= ,
a2 =
6
5
y3 y2 2M2 + M3

= 0.7,
c2 =
1
6

147

d1 = y1 = 0.5.

c1 =

b2 =

9
M2
= ,
2
5

d2 = y2 = 2.0.

Hence the required piecewise cubic splines in each interval are given by
p0 (x) = 0.4x3 + 0.1x,
0x1
3
p1 (x) = (x 1) + 1.2(x 1)2 + 1.3(x 1) + 0.5,
p2 (x) = 0.6(x 2)3 1.8(x 2)2 + 0.7(x 2) + 2.0,

1x2
2 x 3.

Example 3.19.2 Fit a cubic spline curve for the following data with end conditions
y  (0) = 0.2 and y  (3) = 1.
x
y

:
:

0
0

1
0.5

2
3.5

3
5

Solution. Here, the three intervals (0, 1) (1, 2) and (2, 3) are given in each of which
the cubic splines are to be constructed. These cubic spline functions are denoted by
y0 , y1 and y2 . In this example, h1 = h2 = h3 = 1.
For the boundary conditions, equation (3.99) is used. That is,
M0 + 4M1 + M2 = 6(y2 2y1 + y0 )
M1 + 4M2 + M3 = 6(y3 2y2 + y1 ).
Also, from the equations (3.100)
2M0 + M1 = 6(y1 y0 y0 ) and M2 + 2M3 = 6(y3 y3 + y2 )
i.e.,

M0 + 4M1 + M2
M1 + 4M2 + M3
2M0 + M1
M2 + 2M3

=
=
=
=

15
9
1.8
6(1 5 + 3.5) = 15.

## These equations give M0 = 1.36, M1 = 4.52, M2 = 1.72 and M3 = 6.64.

Let the cubic spline in each interval be given by
yi (x) = ai (x xi )3 + bi (x xi )2 + ci (x xi ) + di .

## 148 Numerical Analysis

The coecients are computed as
Mi+1 Mi
Mi
,
,
bi =
6hi+1
2
yi+1 yi 2hi+1 Mi + hi+1 Mi+1
,

ci =
hi+1
6

ai =

di = yi , for i = 0, 1, 2.

Therefore,
M1 M 0
= 0.98, b0 = 0.68, c0 = 0.2, d0 = 0.
6
M2 M 1
= 1.04, b1 = 2.26, c1 = 1.78, d1 = 0.5.
a1 =
6
M3 M 2
= 0.82, b2 = 0.86, c2 = 3.18, d2 = 3.5.
a2 =
6
a0 =

Hence, the required piecewise cubic spline polynomials in each interval are given by
y0 (x) = 0.98x3 0.68x2 + 0.2x,
0x1
3
2
y1 (x) = 1.04(x 1) + 2.26(x 1) + 1.78(x 1) + 0.5,
y2 (x) = 0.82(x 2)3 0.86(x 2)2 + 3.18(x 2) + 3.5,

1x2
2 x 3.

## Example 3.19.3 Consider the function

11
75
3
2

2 x + 26x 2 x + 18, 1 x 2,
f (x) =

## 11 x3 40x2 + 189 x 70, 2 x 3.

2
2
Show that f (x) is a cubic spline.
Solution. Let
11 3
75
x + 26x2 x + 18,
2
2
11 3
189
x 40x2 +
x 70,
and p1 (x) =
2
2
p0 (x) =

1 x 2,
2 x 3.

## Here x0 = 1, x1 = 2 and x2 = 3. The function f (x) will be a cubic spline if

(a)
(b)

pi (xi ) = f (xi ),
pi1 (xi ) = pi (xi ),

pi (xi+1 ) = f (xi+1 ),
pi1 (xi ) = pi (xi ),

i = 0, 1 and
i = 1.

Interpolation

149

But, here the values of f (x0 ), f (x1 ) and f (x2 ) are not supplied, so only the conditions
of (b) are to be checked.
Now,
33
75
33
189
p1 (x) = x2 80x +
p0 (x) = x2 + 52x ,
2
2
2
2
p1 (x) = 33x 80.
p0 (x) = 33x + 52,
p0 (x1 ) = p0 (2) = 0.5, p1 (x1 ) = p1 (2) = 0.5, i.e., p0 (x1 ) = p1 (x1 ).
p0 (x1 ) = p0 (2) = 14 and p1 (x1 ) = p1 (2) = 14. Thus p0 (x1 ) = p1 (x1 ).
Hence f (x) is a spline.
Algorithm 3.5 (Cubic spline). This algorithm nds the cubic spline for each of
the intervals [xi , xi+1 ], i = 0, 1, . . . , n 1. The (i) Natural spline, (ii) Non-periodic
spline or clamped cubic spline, (iii) Extrapolated spline, and (iv) endpoint curvatureadjusted spline are incorporated here. The spacing for xi need not be equal and
assume that x0 < x1 < < xn .
Algorithm Cubic Spline
Read xi , yi , i = 0, 1, 2, . . . , n;//xs are not necessary equispaced//
Read xg; //the value of x at which y is to be computed.//
//Computation of hi .//
for i = 1 to n do
hi = xi xi1 ;
endfor;
//Compututation of Ai , Bi , Ci and Di //
for i = 1 to n 1 do
Bi = 2(hi + hi+1
);
Ci = hi+1 ;
Ai = hi;
yi+1 yi yi yi1
;

Di = 6
hi+1
hi
endfor;
Case :
(i) Natural spline
To nd M1 , M2 , . . . , Mn1 , solve the system of tri-diagonal equation
dened in (3.102).
Set M0 = Mn = 0.
(ii) Non-periodic spline
y at x = x0 , 
xn //
Read y0 , yn ; //rstderivative of 

6 y1 y 0
6
yn yn1


Compute D0 =
y
.
y0 , D n =
h1
h1
hn n
hn
To nd M0 , M1 , M2 , . . . , Mn , solve the system of tri-diagonal equation
dened in (3.106).

## 150 Numerical Analysis

(iii) Extrapolated spline
A1 h1 
A1 h1
, C1 = C1
,
h2
h2
Cn1 hn 
hn Cn1
, Bn1 = Bn1 + Cn1 +
An1 = An1
hn1
hn1
To nd M1 , M2 , . . . , Mn1 , solve the system of tri-diagonal equation
dened in (3.108).
h1 (M2 M1 )
hn (Mn1 Mn2 )
, Mn = Mn1 +
.
Compute M0 = M1
h2
hn1
Read y0 , yn ; //double derivative of y at x = x0 , xn //

= Dn1 Cn1 yn
Compute D1 = D0 A1 y0 , Dn1
To nd M1 , M2 , . . . , Mn1 , solve the system of tri-diagonal equation
dened in (3.109).
Set M0 = y0 , Mn = yn .
endcase;
Compute B1 = A1 + B1 +

## //Compututation of the coecients ai , bi , ci , di , i = 0, 1, 2, . . . , n.//

Mi+1 Mi
for i = 1 to n do
ai =
,
6hi+1
Mi
bi =
2
yi+1 yi 2hi+1 Mi + hi+1 Mi+1
,
d i = yi ;
ci =

hi+1
6
endfor;
//Printing of splines//
for i = 0 to n 1 do
Print Coecient of , i, th spline is ai , bi , ci , di ;
endfor;
//Computation of y at x = xg//
if (xg < x0 ) or (xg > xn ) then
Print x outside the range;
Stop;
endif;
for i = 0 to n 1 do
if (xg < xi+1 ) then
j = i;
exit from for loop;
endif;
endfor;
Compute yc = aj (xg xj )3 + bj (xg xj )2 + cj (xg xj ) + dj ;
Print The value of y at x =, xg, is, yc ;
end Cubic Spline

Interpolation

Program 3.5
.
/* Program Cubic Spline
This program construct cubic splines at each interval
[x[i-1],x[i]], i=1, 2, ..., n and finds the value of
y=f(x) at a given x when the function is supplied as
(x[i],y[i]), i=0, 1, ..., n. */
#include<stdio.h>
#include<math.h>
#include<ctype.h>
#include<stdlib.h>
float M;
void main()
{
int i,n;
char opt,s;
float x,y,h,A,B,C,D;
float a,b,c,d,xg,yd0,ydn,temp,yc;
float TriDiag(float a[],float b[],float c[],float d[],int n);
printf("\nEnter number of subintervals ");
scanf("%d",&n);
printf("Enter x and y values ");
for(i=0;i<=n;i++) scanf("%f %f",&x[i],&y[i]);
printf("Enter interpolating point x ");
scanf("%f",&xg);
printf("The given values of x and y are\nx-value
y-value\n");
for(i=0;i<=n;i++) printf("%f
%f\n",x[i],y[i]);
for(i=0;i<=n;i++) h[i]=x[i]-x[i-1]; /* computation of h[i] */
for(i=1;i<n;i++) /* computation of A,B,C,Ds */
{
A[i]=h[i];
B[i]=2*(h[i]+h[i+1]);
C[i]=h[i+1];
D[i]=6*((y[i+1]-y[i])/h[i+1]-(y[i]-y[i-1])/h[i]);
}
printf("\nN Natural spline\n");
printf("P Non-Periodic spline\n");
printf("E Extrapolated spline\n");
printf("C End point Curvature adjusted spline\n");

151

## 152 Numerical Analysis

printf("
opt=getche();
switch(toupper(opt))
{
case N: /* Natural spline */
temp=TriDiag(A,B,C,D,n-1);
M=0; M[n]=0;
break;
case P: /* Non-periodic spline */
printf("\nEnter the values of y and y[n] ");
scanf("%f %f",&yd0,&ydn);
D=6*((y-y)/h-yd0)/h;
D[n]=6*(ydn-(y[n]-y[n-1])/h[n])/h[n];
for(i=n+1;i>=1;i--) D[i]=D[i-1];
A[n+1]=1; B[n+1]=2;
for(i=n;i>=2;i--){
A[i]=A[i-1]; B[i]=B[i-1]; C[i]=C[i-1];}
B=2; C=1;
temp=TriDiag(A,B,C,D,n+1);
for(i=0;i<=n;i++) M[i]=M[i+1];
break;
case E: /* Extrapolated spline */
B=A+B+A*h/h;
C=C-A*h/h;
A[n-1]=A[n-1]-C[n-1]*h[n]/h[n-1];
B[n-1]=B[n-1]+C[n-1]+C[n-1]*h[n]/h[n-1];
temp=TriDiag(A,B,C,D,n-1);
M=M-h*(M-M)/h;
M[n]=M[n-1]+h[n]*(M[n-1]-M[n-2])/h[n-1];
break;
case C: /* End point Curvature adjusted spline */
printf("\nEnter the values of y and y[n] ");
scanf("%f %f",&yd0,&ydn);
D=D-A*yd0;
D[n-1]=D[n-1]-C[n-1]*ydn;
temp=TriDiag(A,B,C,D,n-1);
M=yd0; M[n]=ydn;
break;

Interpolation

## default : printf("\n No choice \n");

exit(0);
} /* switch */
/* Computation of the coefficients of the splines */
for(i=0;i<=n-1;i++){
a[i]=(M[i+1]-M[i])/(6*h[i+1]); b[i]=M[i]/2;
c[i]=(y[i+1]-y[i])/h[i+1]-(2*h[i+1]*M[i]+h[i+1]*M[i+1])/6;
d[i]=y[i];
}
/* printing of splines */
printf("\nThe cubic splines are \n");
for(i=0;i<n;i++)
{
s=(x[i]>0) ? -:+;
s=(b[i]<0) ? -:+;
s=(c[i]<0) ? -:+;
s=(d[i]<0) ? -:+;
temp=fabs(x[i]);
printf("p%1d(x)=%7.4f(x%c%7.4f)^3%c%7.4f(x%c%7.4f)^2%c%7.4f
(x%c%7.4f)%c%7.4f\n",i,a[i],s,temp,s,fabs(b[i]),
s,temp,s,fabs(c[i]),s,temp,s,fabs(d[i]));
printf("
in [%7.4f,%7.4f]\n",x[i],x[i+1]);
}
/* computation of y at x=xg */
if((xg<x) || (xg>x[n])){
printf("\nx outside the range ");
exit(0);
}
for(i=0;i<=n-1;i++) /* determination of y */
{
if(xg<x[i+1])
{
temp=xg-x[i];
yc=a[i]*temp*temp*temp+b[i]*temp*temp+c[i]*temp+d[i];
printf("The value of y at x=%f is %f ",xg,yc);
exit(0);
}
}
} /* main */

153

## float TriDiag(float a,float b,float c,float d,int n)

{
/* output M[i], i=1, 2,..., n, is a global variable.*/
int i;
float gamma,z;
gamma=b;
for(i=2;i<=n;i++)
{
if(gamma[i-1]==0.0)
{
printf("A minor is zero: Method fails ");
exit(0);
}
gamma[i]=b[i]-a[i]*c[i-1]/gamma[i-1];
}
z=d/gamma;
for(i=2;i<=n;i++)
z[i]=(d[i]-a[i]*z[i-1])/gamma[i];
/* Computation of M */
M[n]=z[n];
for(i=n-1;i>=1;i--)
M[i]=z[i]-c[i]*M[i+1]/gamma[i];
return(M);
} /*end of TriDiag */
A sample of input/output:
Enter number of subintervals 3
Enter x and y values
-1
1.0
1
0.5
2
3.5
3
5.0
Enter interpolating point x 1.2
The given values of x and y are
x-value
y-value
-1.000000
1.000000
1.000000
0.500000
2.000000
3.500000
3.000000
5.000000

Interpolation

N
P
E
C

Natural spline
Non-Periodic spline
Extrapolated spline

## The cubic splines are

p0(x)= 0.3152(x+ 1.0000)^3+ 0.0000(x+ 1.0000)^2- 1.5109(x+ 1.0000)
+ 1.0000 in [-1.0000, 1.0000]
p1(x)=-1.1630(x- 1.0000)^3+ 1.8913(x- 1.0000)^2+ 2.2717(x- 1.0000)
+ 0.5000 in [ 1.0000, 2.0000]
p2(x)= 0.5326(x- 2.0000)^3- 1.5978(x- 2.0000)^2+ 2.5652(x- 2.0000)
+ 3.5000 in [ 2.0000, 3.0000]
The value of y at x=1.200000 is 1.020696
Another input/output:
Enter number of subintervals 3
Enter x and y values
-1
1.0
1
0.5
2
3.5
3
5.0
Enter interpolating point x 1.2
The given values of x and y are
x-value
y-value
-1.000000
1.000000
1.000000
0.500000
2.000000
3.500000
3.000000
5.000000
N
P
E
C

Natural spline
Non-Periodic spline
Extrapolated spline
Enter the values of y and y[n] 0 1
The cubic splines are
p0(x)= 0.6250(x+ 1.0000)^3- 1.3750(x+ 1.0000)^2+ 0.0000(x+ 1.0000)
+ 1.0000 in [-1.0000, 1.0000]

155

## p1(x)=-1.3750(x- 1.0000)^3+ 2.3750(x- 1.0000)^2+ 2.0000(x- 1.0000)

+ 0.5000 in [ 1.0000, 2.0000]
p2(x)= 0.6250(x- 2.0000)^3- 1.7500(x- 2.0000)^2+ 2.6250(x- 2.0000)
+ 3.5000 in [ 2.0000, 3.0000]
The value of y at x=1.200000 is 0.984000

3.20

Bivariate Interpolation

Like single valued interpolation, recently bivariate interpolations become important due
to their extensive uses in a wide range of elds e.g., digital image processing, digital
lter design, computer-aided design, solution of non-linear simultaneous equations etc.
In this section, some of the important methods are described to construct interpolation formulae that can be eciently evaluated.
To construct the formulae, the following two approaches are followed.
(i) Constructing a function that matches exactly the functional values at all the data
points.
(ii) Constructing a function that approximately ts the data. This approach is desirable when the data likely to have errors and require smooth functions.
On the basis of these approaches, one can use four types of methods (i) local matching
methods, (ii) local approximation methods, (iii) global matching methods and (iv) global
approximation methods. In the local methods, the constructed function at any point
depends only on the data at relatively nearby points. In global methods, the constructed
function at any points depends on all or most of the data points.
In the matching method, the matching function matches exactly the given values,
but in the approximate method the function approximately ts the data.
Here, the local and the global matching methods are discussed only.
3.20.1

## Local matching methods

Here two local matching methods are described, viz., triangular interpolation and rectangular grid or bilinear interpolation.
Triangular interpolation
The simplest local interpolating surface is of the form
F (x, y) = a + bx + cy.
The data at the three corners of a triangle determine the coecients. This procedure
generates a piecewise linear surface which is global continuous.

Interpolation

157

Suppose the function f (x, y) be known at the points (x1 , y1 ), (x2 , y2 ) and (x3 , y3 ).
Let f1 = f (x1 , y1 ), f2 = f (x2 , y2 ) and f3 = f (x3 , y3 ).
Let the constructed function be
F (x, y) = a + bx + cy

(3.110)

## such that F (xi , yi ) = f (xi , yi ), i = 1, 2, 3.

Then
f1 = a + bx1 + cy1
f2 = a + bx2 + cy2
f3 = a + bx3 + cy3 .
These equations give the values of a, b and c as
f1 (x2 y3 x3 y2 ) f2 (x1 y3 x3 y1 ) + f3 (x1 y2 x2 y1 )

b =

## (f3 f1 )(x2 x1 ) (f2 f1 )(x3 x1 )

c =

a =

where
= (x2 x1 )(y3 y1 ) (x3 x1 )(y2 y1 ).
The values of a, b, c give the required polynomial.
But, the function F (x, y) can be written in the following form
F (x, y) = Af1 + Bf2 + Cf3 ,
(x2 x)(y3 y) (x3 x)(y2 y)
where A =

B =

## (x1 x)(y2 y) (x2 x)(y1 y)

.
C =

Note 3.20.1

(3.111)
(3.112)
(3.113)
(3.114)

(i) If A + B + C = 1 then = 0.

## (ii) If = 0 then the points (xi , yi ), i = 1, 2, 3 are collinear.

(iii) Let (xi , yi ) and f (xi , yi ), i = 1, 2, . . . , n be given. If we choose non-overlapping
triangles which cover the region containing all these points, then a function that
is continuous in this region can be determined.

## 158 Numerical Analysis

Example 3.20.1 For a function f (x, y), let f (1, 1) = 8, f (2, 1) = 12 and f (2, 2) =
20. Find the approximate value of f (3/2, 5/4) using triangular interpolation.
Solution. Here given that
x1 = 1,
x2 = 2,
x3 = 2,
3
x= ,
2

y1 = 1,
y2 = 1,
y3 = 2,
5
y= .
4

f1 = f (x1 , y1 ) = 8
f2 = f (x2 , y2 ) = 12
f3 = f (x3 , y3 ) = 20

Therefore,
= (x2 x1 )(y3 y1 ) (x3 x1 )(y2 y1 )
= (2 1)(2 1) (2 1)(1 1) = 1.
1
(x2 x)(y3 y) (x3 x)(y2 y)
=
A =

2
1
(x3 x)(y1 y) (x1 x)(y3 y)
=
B =

4
1
(x1 x)(y2 y) (x2 x)(y1 y)
= .
C =

4
Thus f (3/2, 5/4)  F (3/2, 5/4) = Af1 + Bf2 + Cf3 =

1
2

8+

1
4

12 +

1
4

20 = 12.

Bilinear interpolation
Let a function f (x, y) be known at the points (x1 , y1 ), (x1 + h, y1 ), (x1 , y1 + k) and
(x1 + h, y1 + k).
A function F (x, y) is to be constructed within the rectangle formed by these points.
Let f1 = f (x1 , y1 ), f2 = f (x1 + h, y1 ), f3 = f (x1 , y1 + k) and
f4 = f (x1 + h, y1 + k).
Let us construct a function F (x, y) of the form
F (x, y) = a + b(x x1 ) + c(y y1 ) + d(x x1 )(y y1 )

(3.115)

such that
F (x1 , y1 ) = f (x1 , y1 ) = f1 ,

F (x1 + h, y1 ) = f (x1 + h, y1 ) = f2 ,

## F (x1 , y1 + k) = f (x1 , y1 + k) = f3 , F (x1 + h, y1 + k) = f (x1 + h, y1 + k) = f4 .

The unknowns a, b, c, d can be obtained by solving the following equations
f1 = a, f2 = a + bh, f3 = a + ck and f4 = a + hb + kc + hkd.

Interpolation

Thus
a = f1 ,
c =

f2 f1
,
h
f4 + f1 f2 f3
and d =
.
hk

159

b=

f3 f1
k

(3.116)

Example 3.20.2 Find a bilinear interpolation polynomial F (x, y) for the function
f (x, y) where f (1, 1) = 8, f (2, 1) = 10, f (1, 2) = 12 and f (2, 2) = 20. Also, nd an
approximate value of f (4/3, 5/3).
Solution. Here
y1 = 1,
f1 = f (x1 , y1 ) = 8
x1 = 1,
y1 = 1,
f2 = f (x1 + h, y1 ) = 10
x1 + h = 2,
y1 + k = 2,
f3 = f (x1 , y1 + k) = 12
x1 = 1,
y1 + k = 2,
f4 = f (x1 + h, y1 + k) = 20.
x1 + h = 2,
Obviously, h = 1, k = 1.
10 8
f2 f1
Thus,
=
= 2,
a = f1 = 8, b =
h
1
12 8
f4 + f1 f2 f3
f3 f1
=
= 4, d =
= 6.
c=
k
1
hk
Hence,
f (x, y)  F (x, y) = a + b(x x1 ) + c(y y1 ) + d(x x1 )(y y1 )
= 8 + 2(x 1) + 4(y 1) + 6(x 1)(y 1).
38
Therefore,
f (4/3, 5/3) = .
3
3.20.2

Global methods

## Variables-separation method (bilinear form)

Let the interpolating points are evenly distributed over a rectangular grid. Let the
polynomial be of the form
F (x, y) =

n 
n


aij xi y j .

(3.117)

i=1 j=1

## This function can be written as

F (x, y) = [1 y y 2

1
a12 a22 an2 x

y n1 ] .
..
..
.. ..
.
.
.
.
. .
xn1
a1n a2n ann

(3.118)

## 160 Numerical Analysis

That is, F (x, y) = Yt (y)AX(x) (say),
where Y(y) = [1 y y 2 y n1 ]t , X(x) = [1 x x2 xn1 ]t , A = [aij ]nn .
Now, the function F(x, y) is constructed in such a way that
F(x, y) = Y(yj )AX(xi ),
where F is the rearranged array form of the column vector F (xi , yj ), and Y(yj ), X(xi )
are the matrices derived from Yt (y), X(x) by introducing the points (xi , yi ).
That is,

## F (x1 , y1 ) F (x1 , y2 ) F (x1 , yn )

F (x2 , y1 ) F (x2 , y2 ) F (x2 , yn )
,
F=

F (xn , y1 ) F (xn , y2 ) F (xn , yn )

1 x1 x21 x1n1
1 y1 y12 y1n1
n1
2
1 x2 x2 xn1

2
2
, Yt (yj ) = 1 y2 y2 y2 .
X(xi ) =

2
n1
1 xn xn xn
1 yn yn2 ynn1
Since the matrices X, Y and F are known, one can calculate the matrix A as (assuming X and Y are non-singular)
A = (Y1 )t FX1 .

(3.119)

## Thus F (x, y) = Yt (y)A X(x) is the required interpolating polynomial.

Example 3.20.3 Obtain a bilinear interpolating polynomial using the following
data
y
x
1
2

6
10

10
18

Yt (y) = (1 y),
n = 2.

1 1
1 1
t
X(xi ) =
, Y (yj ) =
,
1 2
1 2

1
6 10
1 1
A =
10 18
1 2

2 1
6 10
2 1
2 0
=
=
.
1 1
10 18
1 1
0 4

1 1
1 2

Interpolation
Therefore,

2 0
F (x, y) = [1 y]
0 4

1
x

161

= 2 + 4xy.

## Lagranges bivariate interpolation

Let f (x, y) be a function dened at (m+1)(n+1) distinct points (xi , yi ), i = 0, 1, . . . , m; j =
0, 1, . . . , n. Let us construct a polynomial F (x, y) of degree at most m in x and n in y,
such that
F (xi , yj ) = f (xi , yj ),

i = 0, 1, . . . , m; j = 0, 1, . . . , n.

(3.120)

## As in Lagranges polynomial (3.3) for single variable, we dene

wx (x)
,
(x xi )wx (xi )
wy (y)
Ly,j (y) =
,
(y yj )wy (yj )
Lx,i (x) =

i = 0, 1, . . . , m

(3.121)

j = 0, 1, . . . , n

(3.122)

## where wx (x) = (x x0 )(x x1 ) (x xm ) and wy (y) = (y y0 )(y y1 ) (y yn ).

The functions Lx,i (x) and Ly,j (y) are the polynomials of degree m in x and n in y
respectively and also

0, if xi = xk
0, if yi = yk
and Ly,j (yk ) =
(3.123)
Lx,i (xk ) =
1, if xi = xk
1, if yi = yk
Thus the Lagranges bivariate polynomial is
F (x, y) =

m 
n


(3.124)

i=0 j=0

y
x
0
1

1
1.732051

1.414214
2

## Find the Lagranges bivariate polynomial and hence nd an approximate value of

f (0.25, 0.75).
Solution. m = 1, n = 1, x0 = 0, y0 = 0, x1 = 1, y1 = 1, f (x0 , y0 ) = 1, f (x0 , y1 ) =
1.414214, f (x1 , y0 ) = 1.732051, f (x1 , y1 ) = 2.

Then
F (x, y) =

1
1 


i=0 j=0

## = Lx,0 (x){Ly,0 (y)f (x0 , y0 ) + Ly,1 (y)f (x0 , y1 )}

+Lx,1 (x){Ly,0 (y)f (x1 , y0 ) + Ly,1 (y)f (x1 , y1 )}.
Now,
x1
= 1 x,
Ly,0 (y) = 1 y
01
Lx,1 (x) = x,
Ly,1 (y) = y.
Lx,0 (x) =

Therefore,
F (x, y) = (1 x)(1 y) 1 + (1 x)y 1.414214 + x(1 y) 1.732051 + 2xy
= 1 + 0.732051x + 0.414214y 0.146265xy.
Thus,
F (0.25, 0.75) = 1 + 0.732051 0.25 + 0.414214 0.75 0.146265 0.25 0.75
= 1.466248563.

## Algorithm 3.6 (Lagranges bivariate interpolation). This algorithm nds the

value of f (x, y) by Lagranges interpolation method when a table of values of xi , yj ,
f (xi , yj ), i = 0, 1, . . . , m; j = 0, 1, . . . , n, is given.
Algorithm Lagrange Bivariate
Read xi , i = 0, 1, . . . , m; yj , j = 0, 1, . . . , n;// x and y values //
Read fij , i = 0, 1, . . . , m; j = 0, 1, . . . , n;// fij = f (xi , yj )//
Read xg, yg; //the values of x and y at which f (x, y) is to be determined.//
Set wx = 1, wy = 1;
for i = 0 to m do //computation of wx (xi )//
wx = wx (xg xi );
endfor;
for j = 0 to n do //nd wy (yj )//
wy = wy (yg yj );
endfor;
Set sum = 0;
for i = 0 to m do
for j = 0 to n do

Interpolation

## Compute sum = sum +

wy
wx

fij ;
(xg xi )wdx(i) (yg yj )wdy(j)

endfor;
endfor;
Print The value of f (x, y) is , sum;
end Lagrange Bivariate
function wdx(j)
sum = 0;
for i = 0 to m do
if (i = j) sum = sum + (xj xi );
endfor;
return sum;
end wdx(j)
function wdy(j)
sum = 0;
for i = 0 to n do
if (i = j) sum = sum + (yj yi );
endfor;
return sum;
end wdy(j)
Program 3.6
.
/* Program Lagrange bivariate
This program is used to find the value of a function
f(x,y) at a given point (x,y) when a set of values of
f(x,y) is given for different values of x and y, by
Lagrange bivariate interpolation formula. */
#include<stdio.h>
#include<math.h>
float x,y;
void main()
{
int i,j,n,m;
float xg,yg,f,wx=1,wy=1,sum=0;
float wdx(int j,int m); float wdy(int j,int n);
printf("Enter the number of subdivisions along x and y ");
scanf("%d %d",&m,&n);
printf("Enter x values ");
for(i=0;i<=m;i++) scanf("%f",&x[i]);

163

## printf("Enter y values ");

for(i=0;i<=n;i++) scanf("%f",&y[i]);
printf("Enter function f(x,y) values \n");
for(i=0;i<=m;i++) for(j=0;j<=n;j++)
{printf("f(%f,%f)= ",x[i],y[j]);scanf("%f",&f[i][j]); }
printf("Enter the interpolating point ");
scanf("%f %f",&xg,&yg);
for(i=0;i<=m;i++) wx*=(xg-x[i]);
for(j=0;j<=n;j++) wy*=(yg-y[j]);
for(i=0;i<=m;i++)for(j=0;j<=n;j++)
sum+=wx*wy*f[i][j]/((xg-x[i])*wdx(i,n)*(yg-y[j])*wdy(j,m));
printf("The interpolated value at
(%8.5f,%8.5f) is %8.5f ",xg,yg,sum);
} /* main */
/* function to find w(x[j]) */
float wdx(int j,int m)
{
int i; float prod=1;
for(i=0;i<=m;i++) if(i!=j) prod*=(x[j]-x[i]);
return prod;
}
/* function to find w(y[j]) */
float wdy(int j,int n)
{
int i; float prod=1;
for(i=0;i<=n;i++) if(i!=j) prod*=(y[j]-y[i]);
return prod;
}
A sample of input/output:
Enter the number of subdivisions along x and y 2 2
Enter x values 0 1 2
Enter y values 0 1 2
Enter function f(x,y) values
f(0.000000,0.000000)= 2
f(0.000000,1.000000)= 3
f(0.000000,2.000000)= 6
f(1.000000,0.000000)= 3
f(1.000000,1.000000)= 5

Interpolation

f(1.000000,2.000000)= 9
f(2.000000,0.000000)= 6
f(2.000000,1.000000)= 9
f(2.000000,2.000000)= 14
Enter the interpolating point 0.5 0.5
The interpolated value at ( 0.50000, 0.50000) is

165

2.75000

## Newtons bivariate interpolation formula

Let f (x, y) be dened at (m + 1)(n + 1) distinct points (xi , yj ), i = 0, 1, . . . , m; j =
0, 1, . . . , n. Also, let xs = x0 + sh, yt = y0 + tk, x = x0 + mh and y = y0 + nk.
Some notations are dened in the following:
x f (x, y) = f (x + h, y) f (x, y) = Ex f (x, y) f (x, y)
= (Ex 1)f (x, y)
y f (x, y) = f (x, y + k) f (x, y) = Ey f (x, y) f (x, y)
= (Ey 1)f (x, y)
xx f (x, y) = (Ex2 2Ex + 1)f (x, y) = (Ex 1)2 f (x, y)
yy f (x, y) = (Ey2 1)2 f (x, y)
xy f (x, y) = x {f (x, y + k) f (x, y)}
= {f (x + h, y + k) f (x, y + k)} {f (x + h, y) f (x, y)}
= Ex Ey f (x, y) Ey f (x, y) Ex f (x, y) + f (x, y)
= (Ex 1)(Ey 1)f (x, y)
and so on.
Then,
f (x, y) = f (x0 + mh, y0 + nk) = Exm Eyn f (x0 , y0 )
= (1 + x )m (1 + y )n f (x0 , y0 )


m(m 1)
xx +
= 1 + mx +
2!


n(n 1)
1 + ny +
yy + f (x0 , y0 )
2!

m(m 1)
n(n 1)
xx +
yy
= 1 + mx + ny +
2!
2!

+mnxy + f (x0 , y0 ).
Substituting m =

y y0
x x0
and n =
.
h
k

## 166 Numerical Analysis

x x1
y y1
x x0 h
=
and n 1 =
.
h
h
k

x x0
y y0
x +
y f (x0 , y0 )
F (x, y) = f (x0 , y0 ) +
h
k

2(x x0 )(y y0 )
1 (x x0 )(x x1 )
xy
xx +
+
2!
h2
hk

(y y0 )(y y1 )
+
yy f (x0 , y0 ) +
k2

Then m 1 =
Thus

(3.125)

## which is called Newtons bivariate interpolating polynomial.

Now, introducing unit less quantities u and v dened as x = x0 + uh and y = y0 + vk.
Then x xs = (u s)h and y yt = (v t)k.
Hence, nally (3.125) becomes
1
[u(u 1)xx
2!
+2uvxy + v(v 1)yy ]f (x0 , y0 ) +

## F (x, y) = f (x0 , y0 ) + [ux + vy ]f (x0 , y0 ) +

(3.126)

Example 3.20.5 For the following data obtain Newtons bivariate interpolating
polynomial and hence calculate the values of f (0.75, 0.25) and f (1.25, 1.5).
y
x
0
1
2

1
1
5

3
2
1

5
5
3

Solution.
x f (x0 , y0 ) =
=
y f (x0 , y0 ) =
=
xx f (x0 , y0 ) =
=
yy f (x0 , y0 ) =
=
xy f (x0 , y0 ) =
=

f (x0 + h, y0 ) f (x0 , y0 )
f (x1 , y0 ) f (x0 , y0 ) = 1 1 = 2
f (x0 , y0 + k) f (x0 , y0 )
f (x0 , y1 ) f (x0 , y0 ) = 3 1 = 2
f (x0 + 2h, y0 ) 2f (x0 + h, y0 ) + f (x0 , y0 )
f (x2 , y0 ) 2f (x1 , y0 ) + f (x0 , y0 ) = 5 2 (1) + 1 = 2
f (x0 , y0 + 2k) 2f (x0 , y0 + k) + f (x0 , y0 )
f (x0 , y2 ) 2f (x0 , y1 ) + f (x0 , y0 ) = 5 2 3 + 1 = 0
f (x0 + h, y0 + k) f (x0 , y0 + k) f (x0 + h, y0 ) + f (x0 , y0 )
f (x1 , y1 ) f (x0 , y1 ) f (x1 , y0 ) + f (x0 , y0 ) = 1.

Interpolation

Here h = k = 1. u =
Thus,

167

y y0
x x0
= x, v =
= y.
h
k

F (x, y) = 1 + [x (2) + y 2]
1
+ [x(x 1) (2) + 2xy 1 + y(y 1) 0]
2!
= 1 x + 2y x2 + xy.
Hence f (0.75, 0.25)  F (0.75, 0.25) = 0.375 and f (1.25, 1.5)  F (1.25, 1.5) = 3.0625.

3.21

x2 + 2x + 3
(x + 1)x(x 1)

x
f (x)

:
:

1
2

0
3

1
6

x(x 1)
(x x1 )(x x2 )
=
.
(x0 x1 )(x0 x2 )
2
(x + 1)(x 1)
(x x0 )(x x2 )
=
.
L1 (x) =
(x1 x0 )(x1 x2 )
1
(x + 1)x
(x x0 )(x x1 )
L2 (x) =
=
.
(x2 x0 )(x2 x1 )
2
L0 (x) =

## By Lagranges interpolation formula the polynomial f (x) is given by

x(x 1)
(x + 1)(x 1)
(x + 1)x
2+
3+
6
2
1
2
= x(x 1) 3(x + 1)(x 1) + 3x(x + 1).

f (x) =

Hence

x2 + 2x + 3
f (x)
1
3
3
=
=
+
.
(x + 1)x(x 1)
(x + 1)x(x 1)
x+1 x x1

## Example 3.21.2 Find the missing term in the following table

x
y

:
:

0
1

1
2

2
4

3
?

4
16

Solution.
Method 1.
Using Lagranges formula
(x 1)(x 2)(x 4)
(0 1)(0 2)(0 4)
(x 0)(x 2)(x 4)
L1 (x) =
(1 0)(1 2)(1 4)
(x 0)(x 1)(x 4)
L2 (x) =
(2 0)(2 1)(2 4)
(x 0)(x 1)(x 2)
L3 (x) =
(4 0)(4 1)(4 2)
L0 (x) =

Therefore,

x3 7x2 + 14x 8
.
8
x3 6x2 + 8x
=
.
3
x3 5x2 + 4x
=
.
4
x3 3x2 + 2x
=
.
24
=

x3 6x2 + 8x
x3 7x2 + 14x 8
1+
2
=
8
3
x3 5x2 + 4x
x3 3x2 + 2x
+
4+
16
4
24
5 3 1 2 11
x x + x + 1.
=
24
8
12

## Thus, y(3) = 8.25.

Hence the missing term is 8.25.
Method 2.
Let us construct a polynomial of degree 3 in the form
y(x) = a + bx + cx2 + dx3 .
If the curve passes through the points x = 0, 1, 2, 4, then
a = 1,
a + b + c + d = 2,
a + 2b + 4c + 8d = 4,
a + 4b + 16c + 64d = 16.
Solution of these equations is
a = 1, b =

1
5
11
, c = and d = .
12
8
24

Interpolation

169

Therefore,
y(x) = 1 +

1
11
5
x x2 + x3 .
12
8
24

## Thus y(3) = 8.25.

Example 3.21.3 Let f (x) = log x, x0 = 2 and x1 = 2.1. Use linear interpolation
to calculate an approximate value for f (2.05) and obtain a bound on the truncation
error.
Solution. Let f (x) = log x. The table is
x
y

:
:

2.0
0.693147

2.1
0.741937

## The linear interpolation polynomial is

(x 2.0)
(x 2.1)
0.693147 +
0.741937
(2.0 2.1)
(2.1 2.0)
= 0.487900x 0.282653.

(x) =

## Thus, f (2.05)  (2.05) = 0.717542.

The error in linear interpolation is
  
 f () 
,
|E1 (x)| = |(x x0 )(x x1 )|
2! 

2 2.1.

1
The maximum value of f  (x) = 2 in 2 x 2.1 is |f  (2.0)| = 0.25.
x
Then

0.25 

|E1 (x)| (2.05 2)(2.05 2.1)
 = 0.000313.
2
Thus the upper bound of truncation error is 0.000313.
Example 3.21.4 For the following table nd the value of y at x = 2.5, using
piecewise linear interpolation.
x
y

:
:

1
35

2
40

3
65

4
72

5
80

y(2.5) =

2.5 2
2.5 3
40 +
65 = 52.5.
23
32

## 170 Numerical Analysis

Example 3.21.5 Deduce the following interpolation formula taking three points
x0 , x0 + , 0 and x1 using Lagranges formula.
f (x) =

## (x1 x)(x + x1 2x0 )

(x x0 )(x1 x) 
f (x0 ) +
f (x0 )
(x x0 )2
(x1 x0 )
(x x0 )2
f (x1 ) + E(x)
+
(x1 x0 )2

where
1
E(x) = (x x0 )2 (x x1 )f  () and min{x0 , x0 + , x1 } max{x0 , x0 + , x1 }.
6
Solution. The Lagranges interpolating polynomial for the points x0 , x0 + and x1
is
(x x0 )(x x1 )
(x x0 )(x x1 )
f (x0 ) +
f (x0 + )
f (x) 
(x0 x0 )(x0 x1 )
(x0 + x0 )(x0 + x1 )
(x x0 )(x x0 )
f (x1 ) + E(x)
+
(x1 x0 )(x1 x0 )

(x x0 )(x x1 )
(x x0 )(x x1 )
f (x0 ) +
f (x0 )
=
(x0 x1 )
(x0 + x1 )

+
(x0 x1 + )

(x x0 )(x x0 )
f (x1 ) + E(x)
+
(x1 x0 )(x1 x0 )

## (x x0 )(x x1 ) f (x0 + ) f (x0 )

(2x0 x1 x)(x x1 )
f (x0 ) +
=
(x0 x1 )(x0 x1 + )
(x0 x1 + )

(x x0 )(x x0 )
f (x0 ) + E(x)
+
(x1 x0 )(x1 x0 )
(x x0 )(x1 x) 
(x1 x)(x + x1 2x0 )
f (x0 ) +
f (x0 )
=
2
(x1 x0 )
(x1 x0 )
(x x0 )2
+
f (x1 ) + E(x) as 0.
(x1 x0 )2
The error term is
E(x) = (x x0 )(x x0 )(x x1 )
=

f  ()
3!

1
(x x0 )2 (x x1 )f  (), as 0
6

Interpolation

171

## Example 3.21.6 The standard normal

integral
% xprobability
 1 
2
P (x) =
exp t2 dt
0
2
has the following values
x :
1.00
1.05
1.10
1.15
1.20
1.25
P (x) : 0.682689 0.706282 0.728668 0.749856 0.769861 0.788700
Calculate P (1.235).
Solution. The backward dierence table is
x
1.00
1.05
1.10
1.15
1.20
1.25

P (x)
0.682689
0.706282
0.728668
0.749856
0.769861
0.788700

2 P

3 P

0.023593
0.022386
0.021188
0.020005
0.018839

0.001207
0.001198 0.000009
0.001183 0.000015
0.001166 0.000017

## Here x = 1.235, h = 0.05, xn = 1.25, v = (x xn )/h = (1.235 1.25)/0.05 = 0.3.

v(v + 1) 2
v(v+1)(v+2) 3
P (xn ) +
P (xn )
2!
3!
0.3(0.3 + 1)
(0.001166)
= 0.788700 0.3 0.018839 +
2
0.3(0.3 + 1)(0.3 + 2)
0.000017
+
6
= 0.783169.

## P (1.235) = P (xn ) + vP (xn ) +

Example 3.21.7 Find the seventh and the general terms of the series 3, 9, 20, 38,
65, . . ..
Solution. Let xi = i, i = 1, 2, 3, 4, 5 and y0 = 3, y1 = 9, y2 = 20, y3 = 38, y4 = 65.
We construct the Newtons backward interpolation polynomial using these values.
x
1
2
3
4
5

y
3
9
20
38
65

y 2 y 3 y
6
11
18
27

5
7
9

2
2

## 172 Numerical Analysis

Here xn = 5, v = (x xn )/h = x 5.
v(v + 1) 2
v(v + 1)(v + 2) 3
yn +
yn
2!
3!
v(v + 1)(v + 2)
v(v + 1)
+2
65 + 27v + 9
2!
3!
1
(2v 3 + 33v 2 + 193v + 390)
6
1
[2(x 5)3 + 33(x 5)2 + 193(x 5) + 390]
6
1
(2x3 + 3x2 + 13x).
6

(x) = yn + vyn +
=
=
=
=
The seventh term is

1
(7) = (2 73 + 3 72 + 13 7) = 154.
6
[Other interpolation formulae may also be used to solve this problem.]
Example 3.21.8 From the following table of sin x compute sin 120 and sin 450 .
x
: 100
200
300
400
500
y = sin x : 0.17365 0.34202 0.50000 0.64279 0.76604
Solution. The dierence table is
x
y
0
10 0.17365

2 y

3 y

4 y

0.16837
0.01039

200 0.34202
0.15798
300

0.01519

0.50000
0.14279

0.01954

400 0.64279

0.00480
0.00045
0.00435

0.12325
500 0.76604
(i) To nd sin 120 .
Here x0 = 100 , x = 120 , h = 100 , u =
By Newtons forward formula

(xx0 )
h

(120 100 )
100

= 0.2.

u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0
2!
3!

Interpolation

## u(u 1)(u 2)(u 3) 4

y0
4!
0.2(0.2 1)
(0.01039)
= 0.17365 + 0.2 0.16837 +
2
0.2(0.2 1)(0.2 2)
(0.00480)
+
6
0.2(0.2 1)(0.2 2)(0.2 3)
+
(0.000450)
24
= 0.20791.
+

## (ii) To nd sin 450 .

Here xn = 500 , x = 450 , h = 100 , v = (x xn )/h = (450 500 )/100 = 0.5.
By Newtons backward formula
y(450 ) = sin 450
0.5(0.5 + 1)
= 0.76604 0.5 0.12325 +
(0.01954)
2
0.5(0.5 + 1)(0.5 + 2)
(0.00435)
+
6
0.5(0.5 + 1)(0.5 + 2)(0.5 + 3)
(0.00045)
+
24
= 0.70711.
Example 3.21.9 Use Stirlings formula to nd u32 from the following table
u20 = 14.035, u25 = 13.674, u30 = 13.257,
u35 = 12.734, u40 = 12.089, u45 = 11.309.
Solution. The nite dierence table is shown below.
i xi uxi
uxi
2 20 14.035
0.361
1 25 13.674
0.417
0 30 13.257
0.523
1 35 12.734
0.645
2 40 12.089
0.780
3 45 11.309

2 u x i
0.056
0.106
0.122
0.135

3 u x i

0.050
0.016
0.013

173

## 174 Numerical Analysis

Here x = 32, x0 = 30, h = 5, s = (x x0 )/h = 0.4.
Therefore,
y1 + y0 s2 2
s(s2 1) 3 y2 + 3 y1
+ y1 +
2
2!
3!
2
0.417 0.523 0.42
= 13.257 + 0.4
+
(0.106)
2
2
0.4(0.42 1) 0.050 0.016
+
6
2
= 13.059.

u32 = y0 + s

## Example 3.21.10 The function y =

x is tabulated below.

x : 5600
5700
5800
5900
6000
y : 17.75808 17.86316 17.96702 18.06969 18.17121
Compute

## Solution. The nite dierence table is

i x
y
2 5600 17.75808

2 y

3 y

4 y

0.10508
1 5700 17.86316

0.00122
0.10386

0.00003
0.00119

0 5800 17.96702
0.10267

0.00001
0.00004

0.00115

1 5900 18.06969
0.10152
2 6000 18.17121

(i) For x = 5860, let us take x0 = 5800, then s = (5860 5800)/100 = 0.6.
By Bessels formula
s(s 1) 2 y0 + 2 y1
y 0 + y1
+ (s 0.5)y0 +
2
2!
2
1
3
+ (s 0.5)s(s 1) y1
3!
17.96702 + 18.06969
+ (0.6 0.5) 0.10267
=
2

y(5860) =

Interpolation

## 0.6(0.6 1) 0.00115 0.00119

2!
2
1
+ (0.6 0.5)(0.6)(0.6 1) 0.00004
6
= 18.02877.
+

## (ii) By Stirlings formula

y1 + y0 s2 2
s(s2 1) 3 y2 + 3 y1
+ y1 +
2
2!
3!
2
0.10386 + 0.10267 0.62
= 17.96702 + 0.6
+
(0.00119)
2
2
0.6(0.62 1) 0.00003 + 0.00004
+
6
2
= 18.02877.

y(5860) = y0 + s

Thus

5860 = 18.02877.

Example 3.21.11 Prove that the third order divided dierence of the function
1
1
.
f (x) = with arguments a, b, c, d is
x
abcd
1
.
x
1
1
1
f (b) f (a)
= b a = .
f [a, b] =
ba
ba
ab

## Solution. Here f (x) =

1 + 1
1
f [a, b] f [b, c]
= ab bc =
.
ac
ac
abc
The third order divided dierence is
f [a, b, c] =

f [a, b, c] f [b, c, d]
f [a, b, c, d] =
=

## Example 3.21.12 If f (x) =

1
bcd
1
=
.
abcd

1
abc

1
(1)n
, prove that f [x0 , x1 , . . . , xn ] =
.
x
x0 x1 xn

f (x1 ) f (x0 )
1
(1)1
= x1 x0 =
=
.
x1 x0
x1 x0
x0 x1
x0 x1
1

f [x0 , x1 ] =

175

## 176 Numerical Analysis

The second order divided dierence is
f [x0 , x1 , x2 ] =

1 + 1
f [x0 , x1 ] f [x1 , x2 ])
(1)2
= x0 x1 x1 x2 =
.
x0 x2
x0 x2
x0 x1 x2

## Thus the result is true for n = 1, 2.

Let the result be true for n = k, i.e.,
f [x0 , x1 , . . . , xk ] =
Now,

(1)k
.
x0 x1 xk

## f [x0 , x1 , . . . , xk ] f [x1 , x2 , . . . , xk+1 ]

x0 xk+1

(1)k
1
(1)k
=

x0 xk+1 x0 x1 xk
x1 x2 xk+1


k
1
1
(1)
1

=
x1 x2 xk x0 xk+1 x0 xk+1
(1)k+1
.
=
x0 x1 x2 xk+1

f [x0 , x1 , . . . , xk , xk+1 ] =

## Therefore, the result is true for n = k + 1. Hence by mathematical induction the

result is true for n = 1, 2, . . . .
Example 3.21.13 If f (x) = u(x)v(x) then show that
f [x0 , x1 ] = u(x0 )v[x0 , x1 ] + v(x1 )u[x0 , x1 ]
and hence deduce that if g(x) = w2 (x) then
g[x0 , x1 ] = w[x0 , x1 ][w(x0 ) + w(x1 )].
Solution.

f (x0 ) f (x1 )
u(x0 )v(x0 ) u(x1 )v(x1 )
=
x0 x1
x0 x1
u(x0 )[v(x0 ) v(x1 )] + v(x1 )[u(x0 ) u(x1 )]
=
x0 x1
= u(x0 )v[x0 , x1 ] + v(x1 )u[x0 , x1 ].

f [x0 , x1 ] =

## If g(x) = w2 (x) then let u(x) = v(x) = w(x).

Then
g[x0 , x1 ] = w(x0 )w[x0 , x1 ] + w(x1 )w[x0 , x1 ]
= w[x0 , x1 ][w(x0 ) + w(x1 )].

Interpolation

177

Example 3.21.14 Show that the nth divided dierence f [x0 , x1 , . . . , xn ] can be
expressed as
f [x0 , x1 , . . . , xn ] = D V, where

and


 1
1
1

 x0
x
x
1
2
 2
2
2
 x0
x
x
1
2
D = 

 n1
x
xn1
xn1
1
2
 0
 y0
y1
y2

1
xn
x2n

xnn1
yn















 1
1
1

 x0
x
x
1
2
 2
2
2
 x0
x
x
1
2
V = 

 n1 n1 n1
x
x
x
1
2
 0n
 x
xn1
xn2
0

1
xn
x2n

xnn1
xnn







.






## Solution. The Vandermondes determinant


 1
1
1

 x0
x
x
1
2
 2
 x0
x21
x22

V (x0 , x1 , . . . , xn ) = 

 n1
n1
n1
x
 0 n x1 n x2 n
 x
x1
x2
0
Let


 1

 x0

V (x0 , x1 , . . . , xn1 , x) =  x20

 n
 x0

1
x1
x21

xn1

1
xn
x2n

xn1
n
xnn







.







1 
xn1 x 
x2n1 x2  .

xnn1 xn 
1

## When x = x0 , x1 , . . . , xn1 then V = 0,

i.e., (x x0 ), (x x1 ), (x x2 ), . . . , (x xn1 ) are the factors of V .
Then one can write
V (x0 , x1 , . . . , xn1 , x) = (x x0 )(x x1 ) (x xn1 ),
where is a constant.

## 178 Numerical Analysis

Equating the coecient of xn , the value of is given by


 1 1 1 


 x0 x1 xn1 
 2 2

=  x0 x1 x2n1  = V (x0 , x1 , . . . , xn1 ).
 
 n n

 x0 x1 xnn1 
Therefore,
V (x0 , x1 , . . . , xn ) = V (x0 , x1 , . . . , xn1 )

n1


(xn xi ).

i=0

## Applying this result successively, the explicit expression for V is given by

V (x0 , x1 , . . . , xn ) = {(x0 x1 )(x0 x2 ) (x0 xn )}
{(x1 x2 )(x1 x3 ) (x1 xn )}
{(x2 x3 )(x2 x4 ) (x2 xn )}
{(xn2 xn1 )(xn2 xn )}{(x0 xn )}
n 
n

=
(xi xj )
i=0

j=0
i>j

## Let us consider the determinant


 1
1 1
1 1

 x0
x

x
x
1
i1
i+1 xn
 2
2
2
2
2

x
x

x
x
C(x0 , x1 , . . . , xi1 , xi+1 , . . . , xn ) =  0
1
i1
i+1 xn
 ....................................
 n1 n1
x
x
xn1 xn1 xnn1
0

i1

i+1






.





## It can be shown that

C(x0 , x1 , . . . , xi1 , xi+1 , . . . , xn ) = (1)n V (x0 , x1 , . . . , xi1 , xi+1 , . . . , xn ).
Therefore,
V (x0 , x1 , . . . , xi1 , xi+1 , . . . , xn )
V (x0 , x1 , x2 , . . . , xn )
1
=
(x0 xi )(x1 xi ) (xi1 xi )(xi xi+1 ) (xi xn )
(1)i
.
=
(xi x0 )(xi x1 ) (xi xi+1 )(xi xi+1 ) (xi xn )

Interpolation

179

Now,
D = (1)n y0 C(x1 , x2 , . . . , xn ) + (1)n+1 y1 C(x0 , x2 , . . . , xn ) +
+(1)n+n yn C(x0 , x1 , . . . , xn1 )
n

(1)i yi V (x0 , x1 , . . . , xi1 , xi+1 , . . . , xn ).
=
i=0

Thus,
DV =

n


## (1)i yi V (x0 , x1 , . . . , xi1 , xi+1 , . . . , xn )/V (x0 , x1 , . . . , xn )

i=0

n


(1)i yi

i=0

n

i=0

(1)i
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )

yi
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )

= f [x0 , x1 , . . . , xn ].

## Example 3.21.15 Given y = f (x) in the following table,

x : 10 15 17
y : 3 7 11
Find the values of x when y = 10 and y = 5.
Solution. Here, inverse Lagranges interpolation formula is used in the following
form
n

Li (y)xi .
(y) =
i=0

(y 7)(y 11)
y 2 18y + 77
(y y1 )(y y2 )
=
=
,
(y0 y1 )(y0 y2 )
(3 7)(3 11)
32
(y 3)(y 11)
y 2 14y + 33
(y y0 )(y y2 )
=
=
,
L1 (y) =
(y1 y0 )(y1 y2 )
(7 3)(7 11)
16
(y 3)(y 7)
y 2 10y + 21
(y y0 )(y y1 )
=
=
.
and L2 (y) =
(y2 y0 )(y2 y1 )
(11 3)(11 7)
32
L0 (y) =

## 180 Numerical Analysis

Then
y 2 14y + 33
y 2 10y + 21
y 2 18y + 77
10
15 +
17
32
16
32
1
(137 + 70y 3y 2 ).
=
32

(y) =

1
(137 + 700 300) = 16.78125
32
1
and x(5)  (5) = (137 + 350 75) = 12.87500.
32

Hence,

x(10)  (10) =

## Example 3.21.16 Use inverse Lagranges interpolation to nd a root of the equation y x3 3x + 1 = 0.

Solution. Here y(0) = 1 > 0 and y(1) = 1 < 0. One root lies between 0 and 1.
Now, x and y are tabulated, considering ve points of x as 0, 0.25, 0.50, 0.75 and 1.
x : 0 0.25
0.50
0.75
1.00
y : 1 0.26563 0.37500 0.82813 1.0000
Solution. Here y = 0. Then
L0 (y) =
=
L1 (y) =
L2 (y) =
L3 (y) =
L4 (y) =

## (y y1 )(y y2 )(y y3 )(y y4 )

(y0 y1 )(y0 y2 )(y0 y3 )(y0 y4 )
y1 y 2 y 3 y 4
(y0 y1 )(y0 y2 )(y0 y3 )(y0 y4 )
y0 y2 y3 y4
(y1 y0 )(y1 y2 )(y1 y3 )(y1 y4 )
y0 y1 y3 y4
(y2 y0 )(y2 y1 )(y2 y3 )(y2 y4 )
y0 y1 y2 y4
(y3 y0 )(y3 y1 )(y3 y2 )(y3 y4 )
y0 y1 y2 y3
(y4 y0 )(y4 y1 )(y4 y2 )(y4 y3 )

Therefore,
x  (0) =

4


=
=
=
=
=

0.08249
= 0.02234.
3.69194
0.31054
= 0.47684.
0.65125
0.21997
= 0.88176.
0.24947
0.09961
= 0.63966.
0.15577
0.08249
= 0.30338.
0.27190

i=0

## +0.88176 0.50 + 0.06014 0.75 + 0.30338 1

= 0.38373.
Hence, the approximate root is 0.38373.

Interpolation

3.22

181

Exercise

1. Show that

n

i=1

w(x)
= 1.
(x xi )w (xi )

## 2. Show that L0 (x) + L1 (x) + L2 (x) = 1 for all x.

3. Find a polynomial for f (x) where f (0) = 1, f (1) = 2 and f (3) = 5, using Lagranges method.
4. Show that the truncation error on quadratic interpolation in an equidistant table
is bounded by
 2 
h

## max |f  ()|.

9 3
5. Suppose that f (x) = ex cos x is to be approximated on [0,1] by an interpolating
polynomial on n+1 equally spaced points 0 = x0 < x1 < < xn = 1. Determine
n so that the truncation error will be less than 0.0001 in this interval.
6. Determine an appropriate step size to use, in the construction of a table of f (x) =
(1 + x)4 on [0,1]. The truncation error for linear interpolation is to be bounded
by 5 105 .
% x
dt
is tabulated for equally spaced values of
7. The function dened by f (x) =
1 2 t
x with h = 0.1. What is the maximum error encountered if piecewise quadratic
interpolation is to be used to calculate f (a) where a [1, 2] ?
8. Find the formula for the upper bound of the error involved in linearly interpolating
f (x) between %a and b. Use the formula to nd the maximum error encountered
x

when f (x) =

## et dt is interpolated between x = 0 and x = 1.

9. From the following table, nd the number of students who obtain less than 35
marks
Marks
: 20-30 30-40 40-50 50-60 60-70
No. of Students : 32
53
50
38
33
10. If y(1) = 3, y(3) = 9, y(4) = 30 and y(6) = 132, nd the Lagranges interpolation
polynomial that takes the same values as the function y at the given points.
11. Let the following observation follows the law of a cubic polynomial

## 182 Numerical Analysis

x : 0 1 2 3 4
f (x) : 1 2 1 16 49
Find the extrapolated value of f (5).
12. Use Lagranges interpolation formula to express the function
3x2 + 2x 5
(x 1)(x 2)(x 3)
as sums of partial fractions.
13. Express the function
x3 + 6x + 2
(x + 1)x(x 2)(x 4)
as sums of partial fractions.
14. Using Lagranges interpolation formula, express the function f (x) = 3x2 2x + 5
as the sum of products of the factors (x 1), (x 2) and (x 3) taken two at a
time.
15. Compute the missing values of yn and yn in the following table
yn yn 2 yn

4
5
6
13

18

24

16. The following table gives pressure of a steam plant at a given temperature. Using
Newtons formula, compute the pressure for a temperature of 1420 C.
Temperature 0 C : 140 150 160 170
180
2
Pressure, kgf/cm : 3.685 4.854 6.302 8.076 10.225

Interpolation

183

17. The following data gives the melting point of an alloy of lead and zinc; where T
is the temperature in 0 C and P is the percentage of lead in the alloy. Find the
melting point of the alloy containing 84% of lead using Newtons interpolation
method.
P : 50 60 70 80
T : 222 246 272 299
18. Using a polynomial of third degree, complete the record of the export of a certain
commodity during ve years, as given below.
Year, x
: 1988 1989 1990 1991 1992
Export in tons, y : 450 388 400 470
19. Find the polynomial which attains the following values at the given points.
x : 1 0 1 2 3 4
f (x) : 16 7 4 1 8 29
20. Compute log10 2.5 using Newtons forward dierence interpolation formula, given
that
x
:
2.0
2.2
2.4
2.6
2.8
3.0
log10 x : 0.30103 0.34242 0.38021 0.41497 0.44716 0.47721
21. Find the missing term in the following table:
x : 0 1 2 3 4
y : 1 3 9 81
Why the result diers from 33 = 27 ?
22. In the following table, the value of y are consecutive terms of a series of which the
number 36 is the fth term. Find the rst and the tenth terms of the series. Find
also the polynomial which approximates these values.
x : 3 4 5 6 7 8 9
y : 18 26 36 48 62 78 96
23. From the following table determine (a) f (0.27), and (b) f (0.33).
x : 0.24
0.26
0.28
0.30
0.32
0.34
f (x) : 1.6804 1.6912 1.7024 1.7139 1.7233 1.7532

## 184 Numerical Analysis

24. The population of a town in decennial census were as under. Estimate the population for the year 1955.
Year
: 1921 1931 1941 1951 1961
Population (in core) : 46
68
83
95 105
25. Using Gausss forward formula, nd the value of f (32) given that
f (25) = 0.2707, f (30) = 0.3027, f (35) = 0.3386, f (40) = 0.3794.

26. Using Gausss backward formula, nd the value of 518 given that

## 500 = 22.360680, 510 = 22.583100,

520 = 22.803509, 530 = 23.021729.
27. Use a suitable central dierence formula of either Stirlings or Bessels to nd the
values of f (x) from the following tabulated function at x = 1.35 and at x = 1.42.
x :
1.0
1.2
1.4
1.6
1.8
f (x) : 1.17520 1.50946 1.90430 2.37557 2.94217
28. From Bessels formula, derive the following formula for midway interpolation
1
1
3
(4 y2 + 4 y1 )
y1/2 = (y0 + y1 ) (2 y1 + 2 y0 ) +
2
16
256
Also deduce this formula from Everetts formula.
% /2 
1 sin2 sin2 d is tabulated below:
29. The function log E where E =
0

0
:
0
5
10
15
20
log E : 0.196120 0.195293 0.192815 0.188690 0.182928
Compute log 120 by (a) Bessels formula and (b) Stirlings formula and compare
the results.
30. The value of the elliptic integral
%
E() =

/2

(1 sin2 )1/2 d

for certain equidistance values of are given below. Use Everetts or Bessels
formula to determine E(0.25).

Interpolation

185

:
0.20
0.22
0.24
0.26
0.28
0.30
E() : 1.659624 1.669850 1.680373 1.691208 1.702374 1.713889
31. Using Everetts formula, evaluate f (20) from the following table.
x : 14
18
22
26
f (x) : 2877 3162 3566 3990
32. Using Aitkens method evaluate y when x = 2 from the following table.
x : 1 3 4 6
y : 3 9 30 132
33. Use the Aitkens procedure to determine the value of f (0.2) as accurately as
possible from the following table.
x : 0.17520 0.25386 0.33565 0.42078 0.50946
f (x) : 0.84147 0.86742 0.89121 0.91276 0.93204
34. Show that the rst order divided dierence of a linear polynomial is independent
of the arguments.
35. Show that the second order divided dierence of a quadratic polynomial is constant.
36. If f  (x) is continuous for x0 x x1 , show that f [x0 , x1 ] = f  () where x0
x1 and hence show that
f [x0 , x0 ] lim f [x0 , x1 ] = f  (x0 ).
x1 x0

37. For the equidistant values x0 , x1 , x2 , x3 i.e., xi = x0 + ih, establish the following
relations
1
[f (x1 ) f (x0 )],
h
1
[f (x2 ) 2f (x1 ) + f (x0 )],
f [x0 , x1 , x2 ] =
2!h2
1
[f (x3 ) 3f (x2 ) + 3f (x1 ) f (x0 )].
and f [x0 , x1 , x2 , x3 ] =
3!h3
f [x0 , x1 ] =

38. If f (x) =

ax + b
obtain expressions for f [p, q], f [p, p, q] and f [p, p, q, q].
cx + d

## 186 Numerical Analysis

39. If f (x) = x4 obtain expressions for f [a, b, c], f [a, a, b] and f [a, a, a] where a = b = c.
40. If f (x) = 1/(a x), show that
f [x0 , x1 , . . . , xn ] =

1
.
(a x0 )(a x1 ) (a xn )

## 41. Use Newtons divided dierence interpolation to nd the interpolation polynomial

for the function y = f (x) given by the table:
x : 1 1 4
6
y : 1 3 21 127
42. Use Newtons divided dierence interpolation to nd the interpolating polynomial
for the function y = f (x) given by
x : 1 1 4
6
f (x) : 5 2 26 132
43. Using the given table of value of Bessels function y = J0 (x), nd the root of the
equation J0 (x) = 0 lying in (2.4, 2.6) correct up to three signicant digits.
x : 2.4
2.5
2.6
y : 0.0025 0.0484 0.0968
44. Given below is a table of values of the probability integral
%
2 x x2
y=
e
dx.
0
Determine the value of x for which the value of y is 0.49.
x :
0.45
0.46
0.47
0.48
0.49
y : 0.4754818 0.4846555 0.497452 0.5027498 0.5116683
45. Find x for which cosh x = 1.285, by using the inverse interpolation technique
of successive approximation of Newtons forward dierence interpolation formula,
given the table.
x
:
0.738
0.739
0.740
0.741
0.752
cosh x : 1.2849085 1.2857159 1.2865247 1.2873348 1.2881461

Interpolation

187

46. Use the technique of inverse interpolation to nd x for which sinh x = 5.5 from
the following table.
x : 2.2
2.4
2.6
2.8
sinh x : 4.457 5.466 6.095 8.198
47. Given the following table of f (x) between x = 1.1 and x = 1.5, nd the zero of
f (x).
x : 1.1
1.2
1.3
1.4
1.5
f (x) : 1.769 1.472 1.103 1.344 1.875
48. Use the technique of inverse interpolation to nd a real root of the equation
x3 2x 4 = 0.
49. Using Hermites interpolation formula, estimate the value of log 3.2 from the following table
x
:
3.0
3.5
4.0
y = log x : 1.09861 1.25276 1.38629
: 0.33333 0.28571 0.25000
y  = x1
50. Find the Hermite polynomial of the third degree approximating the function y(x)
such that
y(x0 ) = 1, y(x1 ) = 0 and y  (x0 ) = y  (x1 ) = 0.
51. The following values of x and y are calculated from the relation y = x3 + 10
x : 1 2 3 4 5
y : 11 18 37 74 135
Determine the cubic spline p(x) for the interval [2, 3] given that
(a) p (1) = y  (1) and p (5) = y  (5), (b) p (1) = y  (1) and p (5) = y  (5).
52. Fit a cubic spline to the function dened by the set of points given in the following
table.
x
: 0.10
0.15
0.20
0.25
0.30
x
y = e : 1.1052 1.1618 1.2214 1.2840 1.3499

## 188 Numerical Analysis

Use the end conditions
(a) M0 = MN = 0
(b) p (0.10) = y  (0.10) and p (0.30) = y  (0.30) and
(c) p (0.10) = y  (0.10) and p (0.30) = y  (0.30).
Interpolate in each case for x = 0.12 and state which of the end conditions gives
the best t.
53. The distance di that a car has travelled at time ti is given below.
time ti
: 0 2 4
6
8
distance di : 0 40 160 300 480
Use the values p (0) and p (8) = 98, and nd the clamped spline for the points.
54. Fit a cubic spline for the points (0,1), (1,0), (2,0), (3,1), (4,2), (5,2) and (6,1) and
p (0) = 0.6, p (6) = 1.8 and p (0) = 1 and p (6) = 1.
55. A function f (x) is dened as follows:

1 + x,
0x3
f (x) =
2
1 + x + (x 3) , 3 x 4.
Show that f (x) is a cubic spline in [0, 4].
56. Tabulate the values of the function
f (x, y) = ex sin y + y + 1
for
x = 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5
and y = 0.1, 0.2, 0.3, 0.4, 0.5, 0.6.
Hence nd the value of f (1.6, 0.33) by two-dimensional interpolation.
57. Using the following data obtain the Lagrange and Newtons bivariate interpolating
polynomials
x
: 0 0 0 1 1 1 2 2 2
y
: 0 1 2 0 1 2 0 1 2
f (x, y) : 1 3 7 3 6 11 7 11 17

Chapter 4

## Solution of Algebraic and

Transcendental Equations
Determination of roots of algebraic and transcendental equations is a very important
problem in science and engineering.
A function f (x) is called algebraic if, to get the values of the function starting from
the given value of x, we have to perform arithmetic operations between some real numbers and rational power of x. On the other hand, transcendental functions include
all non-algebraic functions, i.e., an exponential function ex , ax , a logarithmic function
log x, trigonometric functions sin x, cos x, tan x, cot x, etc., inverse trigonometric functions sin1 x, cos1 x, etc. and others.
An equation f (x) = 0 is called algebraic or transcendental according as f (x) is
algebraic or transcendental.
The equations x3 + 7x + 3 = 0, x5 7x2 + 3 = 0 are algebraic equations where as
x
e + sin x = 0, 5 log x + 3x 2 = 0 are the transcendental equations.
The denition of roots of an equation can be given in two dierent ways:
Algebraically, a number c is called a root of the equation f (x) = 0 i f (c) = 0 and
geometrically, the real roots of the equation f (x) = 0 are the values of x where the
graph of y = f (x) meets the x-axis.
Development of numerical methods to solve algebraic or transcendental equations are
very much essential because the analytic method fails to solve the polynomial equations
of degree greater than four.
Most of the numerical methods, used to solve an equation are based on iterative
techniques. Dierent numerical methods are available to solve the equation f (x) =
0. But, each method has some advantages and disadvantages over another method.
Generally, the following aspects are considered to compare the methods:
189

## 190 Numerical Analysis

convergence or divergence, rate of convergence, applicability of the method, amount of
pre-calculation needed before application of the method, etc.
The process of nding the approximate values of the roots of an equation can be
divided into two stages: (i) location of the roots, and (ii) computation of the values of
the roots with the specied degree of accuracy.

4.1

Location of Roots

An interval [a, b] is said to be the location of a real root c if f (c) = 0 for a < c < b.
Mainly, two methods are used to locate the real roots of an equation, one is graphical
method and other is an analytic method known as method of tabulation.
4.1.1

Graphical method

First method:
In this method, the graph of y = f (x) is drawn in a rectangular co-ordinates system.
It is obvious that the abscissas of the points where the graph intersects the x-axis are
the roots of the equation f (x) = 0. But, practically, it is most dicult to determine
the exact value of x where the graph intersects the x-axis. For example, if x = 1.27831
is a root of an equation f (x) = 0 then we can not determine 1.2783 (four digits after
decimal point) from the graph. We can measure the value of x up to one or two decimal
places. But, the approximate value of the root can be determined using this method.
Second method:
Some times, the approximate roots of f (x) = 0 can be determined by dividing all
the terms of the equation into groups, one of them is written on the left-hand side of
the equation and the other on the right hand side, i.e., the equation is represented as
g(x) = h(x). Then the graph of two functions y = g(x) and y = h(x), are drawn. The
abscissas of the points of intersection of these graphs are the roots of the equation.
Example 4.1.1 Use graphical method to locate the roots of the equation x3 4x
2 = 0.
Solution. First method:
The graph of the function y = x3 4x 2 is drawn in Figure 4.1(a). The curve cuts
the x-axis at three points and, consequently, the equation has three real roots. From
gure, it is observed that the roots belong to the intervals [2, 1], [1, 0] and [2, 3].
Second method:
The given equation can be written as x3 = 4x + 2. The graph of the functions y = x3
and y = 4x + 2 are drawn (Figure 4.1(b)). The abscissas of the points of intersection
of the graphs of these functions are roots of the equations. The intervals of the roots
are [2, 1], [1, 0] and [2, 3].

f (x)

yM = 4x + 2

191

yI
= x3
O
-2

-1

- x

## (b) The graph of y = x3 and y = 4x + 2.

(a) The graph of y = x3 4x 2.
Figure 4.1: Illustration of location of roots.
The graphical method to locate the roots is not very useful, because, the drawing of
the function y = f (x) is itself a complicated problem. But, it makes possible to roughly
determine the intervals to locate the roots. Then an analytic method is used to locate
the root.
4.1.2

Method of tabulation

This method depends on the continuity of the function f (x). Before applying the
tabulation method, following result should be noted.
Theorem 4.1 If f (x) is continuous in the interval (a, b) and if f (a) and f (b) have the
opposite signs, then at least one real root of the equation f (x) = 0 lies within the interval
(a, b).
If f (a) and f (b) have same signs then f (x) = 0 has no real roots or has an even
number of real roots.
If the curve y = f (x) touches the x-axis at some point, say, x = c then c is a root
of f (x) = 0 though f (a) and f (b), a < c < b may have same sign. For example,
f (x) = (x 2)2 touches the x-axis at x = 2, also f (1.5) > 0 and f (2.5) > 0, but, x = 2
is the root of the equation f (x) = (x 2)2 = 0.
A trial method for tabulation is as follows:
Form a table of signs of f (x) setting x = 0, 1, 2, . . .. If the sign f (x) changes it signs
for two consecutive values of x then at least one root lies between these two values, i.e.,
if f (a) and f (b) have opposite signs then a root lies between a and b.
Example 4.1.2 Find the location of roots of the equation 8x3 20x2 2x + 5 = 0
by tabulation method.

## 192 Numerical Analysis

Solution. We form a table of sign of f (x), taken x = 0, 1, 1, 2, 2, . . . as follows:
x
0 1 1 2 2 3
Sign of f (x) + +
The equation has three real roots as its degree is 3. The location of the roots of the
given equation are (1, 0), (0, 1) and (2, 3).
A systematic process for tabulation
The following sequence of steps to be performed to locate the roots of an equation
f (x) = 0 by tabulation method:
1. nd the rst derivative f  (x),
2. prepare a table of signs of the function f (x) by setting x equal to
(a) the roots of f  (x) = 0 or the values close to them,
(b) the boundary values (preceding from the domain of permissible values of
the variable),
3. determine the intervals at the endpoints of which the function assumes values
of opposite signs. These intervals contain one and only one root each in its
interior.

Example 4.1.3 Find the number of real roots of the equation 3x 3x 2 = 0 and
locate them.
Solution. Let f (x) = 3x 3x 2. The domain of denition of the function f (x) is
(, ).
Now, f  (x) = 3x log 3 3.
The roots of f  (x) = 0 is given by 3x log 3 3 = 0
log 3 log log 3
3
or, x =
= 0.914.
log 3
log 3
A table of signs of f (x) is then form by setting x equal to
(a) the values close of the roots of f  (x) = 0, i.e., x = 0, x = 1 and
(b) boundary values of domain, i.e., x = .
or, 3x =

x
0 1
Sign of f (x) + +
The equation 3x 3x 2 = 0 has two real roots since the function twice changes sign,
among them one is negative root and other is greater than 1.

## Sol. of Algebraic and Transcendental Equs.

193

A new table with small intervals of the location of the root is constructed in the
following.
x
0 1 1 2
Sign of f (x) + +
The roots of the given equation are in (1, 0) [as f (0).f (1) < 0] and (1, 2).
This section is devoted to locate the roots which is the rst stage of solution of
algebraic and transcendental equations.
The second stage is the computation of roots with the specied degree of accuracy. In
the following sections some methods are discussed to determine the roots of an algebraic
or a transcendental equation. Before presenting the solution methods we dene the order
of convergence of a sequence of numbers in the following.
Order of Convergence
Assume that the sequence {xn } of numbers converges to and let n = xn for n 0.
If two positive constants A = 0 and p > 0 exist and
lim

n+1
=A
pn

(4.1)

then the sequence is said to converge to with order of convergence p. The number A
is called the asymptotic error constant.
If p = 1, the order of convergence of {xn } is called linear and if p = 2, the order of
In the next section, one of the bracketing method called bisection method is introduced.

4.2

Bisection Method

Let be a root of the equation f (x) = 0 lies in the interval [a, b], i.e., f (a).f (b) < 0,
and (b a) is not suciently small. The interval [a, b] is divided into two equal intervals
a+b
ba
, and c =
(Figure 4.2). If f (c) = 0, then c is
[a, c] and [c, b], each of length
2
2
an exact root.
Now, if f (c) = 0, then the root lies either in the interval [a, c] or in the interval [c, b].
If f (a).f (c) < 0 then the interval [a, c] is taken as new interval, otherwise [c, b] is taken
as the next interval. Let the new interval be [a1 , b1 ] and use the same process to select
the next new interval. In the next step, let the new interval be [a2 , b2 ]. The process of
bisection is continued until either the midpoint of the interval is a root, or the length
(bn an ) of the interval [an , bn ] (at nth step) is suciently small. The number an and

## bn are the approximate roots of the equation f (x) = 0. Finally, xn =

as the approximate value of the root .

an + bn
is taken
2

a
O

- x

## Figure 4.2: Illustration of bisection method.

It may be noted that when the reduced interval be [a1 , b1 ] then the length of the
interval is (b a)/2, when the interval be [a2 , b2 ] then the length is (b a)/22 . At the
an + bn
nth step the length of the interval being (b a)/2n . In the nal step, when =
2
is chosen as a root then the length of the interval being (b a)/2n+1 and hence the error
does not exceed (b a)/2n+1 .
Thus, if be the error at the nth step then the lower bound of n is obtained from
the following relation
|b a|
.
2n
The lower bound of n is obtained by rewriting this inequation as
n

.
log 2

(4.2)

(4.3)

## Hence the minimum number of iterations required to achieve the accuracy is



loge |ba|

.
(4.4)
log 2
For example, if the length of the interval is |b a| = 1 and = 0.0001, then n is given
by n 14.
The minimum number of iterations required to achieved the accuracy for |b a| = 1
are shown in Table 4.1.
Theorem 4.2 Assume that f (x) is a continuous function on [a, b] and that there exists
a number [a, b] such that f () = 0. If f (a) and f (b) have opposite signs, and {xn }
represents the sequence of midpoints generated by the bisection method, then
| xn |

ba
for n = 0, 1, 2, . . .
2n+1

(4.5)

195

## Table 4.1: Number of iterations for given .

102 103 104 105 106 107
n 7
10
14
17
20
24

## and therefore the sequence {xn } converges to the root i.e.,

lim xn = .

Proof. The root and the midpoint xn both lie in the interval [an , bn ], the distance
between xn and cannot be greater than half the width of the interval [an , bn ]. Thus
| xn |

|bn an |
for all n.
2

(4.6)

From the bisection method, it is observed that the successive interval widths form
the following pattern.
|b0 a0 |
, where b0 = b and a0 = a,
21
|b0 a0 |
|b1 a1 |
=
,
|b2 a2 | =
2
22
|b2 a2 |
|b0 a0 |
.
=
|b3 a3 | =
2
23
|b1 a1 | =

Hence

|b0 a0 |
.
2n
| xn |

|b0 a0 |
[using (4.6)].
2n+1

## Now, the limit gives

| xn | 0 as n

i.e., lim xn = .
n


Note 4.2.1 If the function f (x) is continuous on [a, b] then the bisection method is
applicable. This is justied in Figure 4.3. For the function f (x) of the graph of Figure
4.3, f (a) f (b) < 0, but the equation f (x) = 0 has no root between a and b as the
function is not continuous at x = c.

## 196 Numerical Analysis

f (x)
6

b
O

- x

Figure 4.3: The function has no root between a and b, though f (a) f (b) < 0.
Note 4.2.2 This method is very slow, but it is very simple and will converge surely
to the exact root. So the method is applicable for any function only if the function is
continuous within the interval [a, b], where the root lies.
In this method derivative of the function f (x) and pre-manipulation of function are
not required.
Note 4.2.3 This method is also called bracketing method since the method successively
reduces the two endpoints (brackets) of the interval containing the real root.
Example 4.2.1 Find a root of the equation x2 + x 7 = 0 by bisection method,
correct up to two decimal places.
Solution. Let f (x) = x2 + x 7.
f (2) = 1 < 0 and f (3) = 5 > 0. So, a root lies between 2 and 3.
n
0
1
2
3
4
5
6
7
8
9
10

bn
xn+1
an
2
3
2.5
2
2.5
2.250
2
2.250
2.125
2.125
2.250
2.188
2.188
2.250
2.219
2.188
2.219
2.204
2.188
2.204
2.196
2.188
2.196
2.192
2.192
2.196
2.194
2.192
2.194
2.193
2.192
2.193
2.193

f (xn+1 )
1.750
0.313
-0.359
-0.027
0.143
0.062
0.018
-0.003
0.008
0.002
0.002

## Sol. of Algebraic and Transcendental Equs.

197

Algorithm 4.1 (Bisection method). This algorithm nds a real root of the equation f (x) = 0 which lies in [a, b] by bisection method.
Algorithm Bisection
Input function f (x);
// Assume that f (x) is continuous within [a, b] and a root lies on [a, b].//
//tolerance for width of the interval//
//input of the interval//
Compute f a = f (a); f b = f (b);
//compute the function values//
if sign(f a) = sign(f b) then
//sign(f a) gives the sign of the value of f a.//
Print f (a) f (b) > 0, so there is no guarantee for a root within [a, b];
Stop;
endif;
do
Compute c = (a + b)/2;
Compute f c = f (c);
if f c = 0 or |f c| < then
a = c and b = c;
else if sign(f b) = sign(f c) then
b = c; f b = f c;
else
a = c; f a = f c;
endif;
while (|b a| > );
Print the desired root is c;
end Bisection
Program 4.1
.
/* Program Bisection
Program to find a root of the equation x*x*x-2x-1=0 by
bisection method.
Assume that a root lies between a and b. */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
#define f(x) x*x*x-2*x-1 /* definition of the function f(x) */
void main()
{
float a,b,fa,fb,c,fc;
float eps=1e-5; /* error tolerance */

## 198 Numerical Analysis

printf("\nEnter the value of a and b ");
scanf("%f %f",&a,&b);
fa=f(a); fb=f(b);
if(fa*fb>0)
{
printf("There is no guarantee for a root within [a,b]");
exit(0);
}
do
{
c=(a+b)/2.;
fc=f(c);
if((fc==0) || (fabs(fc)<eps))
{
a=c;b=c;
}
else if(fb*fc>0)
{
b=c; fb=fc;
}
else
{
a=c; fa=fc;
}
}while(fabs(b-a)>eps);
printf("\nThe desired root is %8.5f ",c);
} /* main */
A sample of input/output:
Enter the value of a and b 0 2
The desired root is 1.61803
Another popular method is the method of false position or the regula falsi method.
This is also a bracketing method. This method was developed because the bisection
method converges at a fairly slow speed. In general, the regula falsi method is faster
than bisection method.

4.3

## Regula-Falsi Method (Method of False Position)

The Regula-Falsi method is one of the most widely used methods of solving algebraic
and transcendental equations. This method is also known as method of false position,

199

## method of chords and the method of linear interpolation.

Let a root of the equation f (x) = 0 be lies in the interval [a, b], i.e., f (a) f (b) < 0.
The idea of this method is that on a suciently small interval [a, b] the arc of the curve
y = f (x) is replaced by the chord joining the points (a, f (a)) and (b, f (b)). The abscissa
of the point of intersection of the chord and the x-axis is taken as the approximate value
of the root.
Let x0 = a and x1 = b. The equation of the chord joining the points (x0 , f (x0 )) and
(x1 , f (x1 )) is
x x0
y f (x0 )
=
.
f (x0 ) f (x1 )
x0 x1

(4.7)

To nd the point of intersection, set y = 0 in (4.7) and let (x2 , 0) be such point.
Thus,
x2 = x0

f (x0 )
(x1 x0 ).
f (x1 ) f (x0 )

(4.8)

This is the second approximation of the root. Now, if f (x2 ) and f (x0 ) are of opposite
signs then the root lies between x0 and x2 and then we replace x1 by x2 in (4.8). The
next approximation is obtained as
x3 = x0

f (x0 )
(x2 x0 ).
f (x2 ) f (x0 )

If f (x2 ) and f (x1 ) are of opposite signs then the root lies between x1 and x2 and the
new approximation x3 is obtain as
x3 = x2

f (x2 )
(x1 x2 ).
f (x1 ) f (x2 )

The procedure is repeated till the root is obtained to the desired accuracy.
If the nth approximate root (xn ) lies between an and bn then the next approximate
root is thus obtained as
xn+1 = an

f (an )
(bn an ).
f (bn ) f (an )

(4.9)

## The illustration of the method is shown in Figure 4.4.

This method is also very slow and not suitable for hand calculation. The advantage
of this method is that it is very simple and the sequence {xn } is sure to converge. The
another advantage of this method is that it does not require the evaluation of derivatives
and pre-calculation.

f (x)

f (x)

x0 =a
O

x1 x2 s x
b

x0

x1x2

- x

## Figure 4.4: Illustration of Regula-falsi method.

To estimate the error of approximation, the following formula may be used
| xn | < |xn xn1 |

(4.10)

where is an exact root and xn1 and xn are its approximations obtained at the (n1)th
and nth iterations. This relation can be used when
M 2m, where M = max |f  (x)| and m = min |f  (x)| in [a, b].

(4.11)

## Example 4.3.1 Find a root of the equation x3 + 2x 2 = 0 using Regula-Falsi

method, correct up to three decimal places.
Solution. Let f (x) = x3 + 2x 2. f (0) = 2 < 0 and f (1) = 1 > 0. Thus, one root
lies between 0 and 1. The calculations are shown in the following table.
n
0
1
2
3
4

left end
point an
0.0000
0.6700
0.7570
0.7690
0.7707

right end
point bn
1.0
1.0
1.0
1.0
1.0

f (an )
2.0000
0.3600
0.0520
0.0072
0.0010

f (bn )
1.0
1.0
1.0
1.0
1.0

xn+1
0.6700
0.7570
0.7690
0.7707
0.7709

f (xn+1 )
0.3600
0.0520
0.0072
0.0010
0.0001

## Therefore, a root of the equation is 0.771 correct up to three decimal places.

Algorithm 4.2 (Regula-Falsi). This algorithm nds a root of the equation f (x) =
0 which lies in [x0 , x1 ], by Regula-Falsi method.
Algorithm Regula-Falsi
Input function f (x);

201

## Read x0 , x1 , ; //interval for the root and error tolerance//

Compute f x0 = f (x0 ); f x1 = f (x1 ); // Compute the function values at x0 and x1 //
do
x0 f x1 x1 f x0
;
Compute x2 =
f x1 f x0
Compute f x2 = f (x2 );
if |f x2 | then
Print The root is, x2 ;
Stop;
endif;
if sign(f x2 ) = sign(f x0 ) then
Set x1 = x2 ; f x1 = f x2 ;
else
Set x0 = x2 ; f x0 = f x2 ;
endif;
while (|f x2| > );
end Regula-Falsi
Program 4.2
.
/* Program Regula-Falsi
Program to find a root of the equation x*x-2x-3=0 by regula
falsi method. Assumed that a root lies between x0 and x1. */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
#define f(x) x*x-2*x-3 /* definition of the function f(x) */
void main()
{
float x0,x1,x2,fx0,fx1,fx2;
float eps=1e-5; /* error tolerance */
printf("\nEnter the value of x0 and x1 ");
scanf("%f %f",&x0,&x1);
fx0=f(x0); fx1=f(x1);
if(fx0*fx1>0)
{
printf("There is no guarantee for a root within
[%6.3f,%6.3f]",x0,x1);
exit(0);
}

## 202 Numerical Analysis

do
{
x2=(x0*fx1-x1*fx0)/(fx1-fx0);
fx2=f(x2);
if(fabs(fx2)<eps)
{
printf("The root is %8.5f ",x2);
exit(0);
}
if(fx2*fx0<0)
{
x1=x2; fx1=fx2;
}
else
{
x0=x2; fx0=fx2;
}
}while(fabs(fx2)>eps);
} /* main */
A sample of input/output:
Enter the value of x0 and x1 0 3
The root is 3.00000

4.4

## The iteration method or the method of successive approximations, is one of

the most important methods in numerical mathematics. This method is also known as
xed-point iteration.
Let f (x) be a function continuous on the interval [a, b] and the equation f (x) = 0
has at least one root on [a, b]. The equation f (x) = 0 can be written in the form
x = (x).

(4.12)

Suppose x0 [a, b] be an initial guess to the desired root . Then (x0 ) is evaluated
and this value is denoted by x1 . It is the rst approximation of the root . Again, x1 is
substituted for x to the right side of (4.12) and obtained a new value x2 = (x1 ). This
process is continued to generate the sequence of numbers x0 , x1 , x2 , . . . , xn , . . ., those
are dened by the following relation:
xn+1 = (xn ),

n = 0, 1, 2, . . .

(4.13)

## Sol. of Algebraic and Transcendental Equs.

203

This successive iterations are repeated till the approximate numbers xn s converges
to the root with desired accuracy, i.e., |xn+1 xn | < , where is a suciently small
number. The function (x) is called the iteration function.
Note 4.4.1 There is no guarantee that this sequence x0 , x1 , x2 , . . . will converge. The
function f (x) = 0 can be written as x = (x) in many dierent ways. This is very
signicant since the form of the function (x) is very important both for the convergence
and for its rate.
For example, the equation x3 + x2 1 = 0 has a root lies between 0 and 1. This
equation can be rewritten in the following ways:


1
1 x2
1 x2
2 1/3
, etc.
;
x
=
;
x
=
(1

x
)
;
x
=
1 x3 ; x =
x=
2
x
x
1+x
The following theorem gives the sucient condition for convergence of the iteration
process.
Theorem 4.3 Let be a root of the equation f (x) = 0 and it can be written as x = (x)
and further that
1. the function (x) is dened and dierentiable on the interval [a, b],
2. (x) [a, b] for all x [a, b],
3. there is a number l < 1 such that
| (x)| l < 1 for x [a, b].

(4.14)

Then the sequence {xn } given by (4.13) converges to the desired root irrespective of
the choice of the initial approximation x0 [a, b] and the root is unique.
Proof. Since is a root of the equation x = (x), therefore
= ().

(4.15)

xi+1 = (xi ).

(4.16)

## Subtracting (4.16) from (4.15),

xi+1 = () (xi )
= ( xi ) (i ) (by mean value theorem)
(where i lies between and xi )

## 204 Numerical Analysis

For i = 0, 1, 2, . . . , n,
x1 = ( x0 ) (0 ), 0 lies between and x0
x2 = ( x1 ) (1 ), 1 lies between and x1
..
..
.
.
xn+1 = ( xn ) (n ), n lies between and xn .
Multiplying all these equations we obtain the following result
( xn+1 ) = ( x0 ) (0 ) (1 )  (n )
or | xn+1 | = | x0 || (0 )|| (1 )| | (n )|
ln+1 | x0 |,

(4.17)

## where | (x)| l for all x [a, b].

Now, if l < 1 then the right hand side of (4.17) tends to zero as n .
Therefore, lim xn+1 = .
n

Hence the sequence {xn } converge to if | (x)| < 1, for all x [a, b].
Now to prove the uniqueness.
Let 1 and 2 be two roots of x = (x), i.e., 1 = (1 ) and 2 = (2 ). Then
|2 1 | = |(2 ) (1 )| = | (c)||1 2 |,

(4.18)

where c (1 , 2 ).
Equation (4.18) reduces to
|1 2 |(1 | (c)|) = 0
and by condition (iii) 1 = 2 , i.e., the two roots are not distinct, they are equal.
4.4.1

Estimation of error

## Let be an exact root of the equation x = (x) and xn+1 = (xn ).

Therefore,
| xn | = |() (xn1 )| = | xn1 | | (c)|, c lies between xn1 and
l| xn1 |
= l| xn + xn xn1 | l| xn | + l|xn xn1 |.
After rearrangement, this relation becomes
| xn |

l
ln
|xn xn1 |
|x1 x0 |.
1l
1l

(4.19)

## Sol. of Algebraic and Transcendental Equs.

205

Let the maximum number of iterations needed to achieve the accuracy be N ().
Thus from (4.19)
lN
|x1 x0 | .
1l
This gives
log |x(1l)
1 x0 |
.
(4.20)
N ()
log l
For l

1
2

## the estimation of the error is given by the following simple form:

| xn | |xn xn1 |.

(4.21)

Order of convergence
The convergence of an iteration method depends on the suitable choice of the iteration
function (x) and x0 , the initial guess.
Let xn converges to the exact root , so that = ().
Thus xn+1 = (xn ) ().
Let n+1 = xn+1 . Note that  (x) = 0. Then the above relation becomes
n+1 = (n + ) ()
1
= n  () + 2n  () +
2
= n  () + O(2n )
i.e.,

n+1  n  ().

## Hence the order of convergence of iteration method is linear.

Geometric interpretation
Geometrically, the point of intersection of the line y = x and the curve y = (x) is a
root of the equation f (x) = 0. Depending on the value of  () the convergence and
divergence cases are illustrated in Figures 4.5-4.6.
Merit and Demerit
The disadvantage of iteration method is that a pre-calculation is required to rewrite
f (x) = 0 into x = (x) in such a way that | (x)| < 1. But, the main advantage of this
method is that the operation carried out at each stage are of the same kind, and this
makes easier to develop computer program.
This method is some times called a linear iteration due to its linear order of convergence.

## 206 Numerical Analysis

f (x)

f (x)

y=x

y=x


?

6 -66
6
?

??


?
6

y = (x)


?

y = (x)

- x
- x
x0x2 x3x1
x1
x0
x2

O
O
(b) Spiral case solution,
(a) Stair case solution,
1 <  () < 0.
0 <  () < 1
Figure 4.5: Convergent for | ()| < 1.

f (x)
6

y=(x)

f (x)

y=x
6

y=x

6
6





x0

x1

x
O

x3x1

x0x2 x4

y = (x)
-

## Figure 4.6: Divergent for | ()| > 1.

Example 4.4.1 Consider the equation 5x3 20x + 3 = 0. Find the root lying on
the interval [0,1] with an accuracy of 104 .
3

## Solution. The given equation is written as x = 5x20+3 = (x) (say).

2
3x2
Now,  (x) = 15x
20 = 4 < 1 on [0,1]. Let x0 = 0.5. The calculations are shown in
the following table.
n
0
1
2
3
4

xn
(xn ) = xn+1
0.5
0.18125
0.18125
0.15149
0.15149
0.15087
0.15087
0.15086
0.15086
0.15086

## Sol. of Algebraic and Transcendental Equs.

207

At this stage the iteration process is terminated and = 0.1509 is taken as the
required root.
Example 4.4.2 Find a root of the equation
cos x xex = 0
correct up to three decimal places.
Solution. It is easy to see that one root of the given equation lies between 0 and 1.
Let x0 = 0. The equation can be written as x = ex cos x = (x) (say).
The calculations are shown in the following table.
n
0
1
2
3
4
5
6
xn 0.50000 0.53228 0.50602 0.52734 0.51000 0.52408 0.51263
xn+1 0.53228 0.50602 0.52734 0.51000 0.52408 0.51263 0.52193
n
7
8
9
10
11
12
13
xn 0.52193 0.51437 0.52051 0.51552 0.51958 0.51628 0.51896
xn+1 0.51437 0.52051 0.51552 0.51958 0.51628 0.51896 0.51678
n
14
15
16
17
18
19
20
xn 0.51678 0.51855 0.51711 0.51828 0.51733 0.51810 0.51748
xn+1 0.51855 0.51711 0.51828 0.51733 0.51810 0.51748 0.51798
Therefore, the required root is 0.518 correct up to three decimal places.
Algorithm 4.3 (Fixed point iteration). This algorithm computes a root of the
equation f (x) = 0 by rewriting the equation as x = (x), provided | (x)| < 1 in the
interval [a, b], by xed point iteration method. x0 [a, b] be the initial guess and
is the error tolerance.
Algorithm Iteration
Input function (x);
Read x0 , ; //initial guess and error tolerance.//
Set x1 = x0 ;
do
Set x0 = x1 ;
Compute x1 = (x0 );
while (|x1 x0 | > );
Print The root is, x1 ;
end Iteration

## 208 Numerical Analysis

Program 4.3
.
/* Program Fixed-Point Iteration
Program to find a root of the equation x*x*x-3x+1=0
by fixed point iteration method. phi(x) is obtained
by rewrite f(x)=0 as x=phi(x), which is to be supplied.*/
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
#define phi(x) (3*x-1)/(x*x)
/* definition of the function phi(x) and it to be
changed accordingly */
void main()
{
int k=0; /* counts number of iterations */
float x1,x0; /* initial guess */
float eps=1e-5; /* error tolerance */
printf("\nEnter the initial guess x0 ");
scanf("%f",&x0);
x1=x0;
do
{
k++;
x0=x1;
x1=phi(x0);
}while(fabs(x1-x0)>eps);
printf("One root is %8.5f obtained at %d th iteration ",x1,k);
} /* main */
A sample of input/output:
Enter the initial guess x0 1
One root is 1.53209 obtained at 37th iteration

4.5

## Acceleration of Convergence: Aitkens 2 -Process

The rate of convergence of iteration method is linear. But, this slow rate can be accelerated by using Aitkens method.
The iteration scheme of this method is obtained from xed point iteration method as
xn+1 = (xn ) with | (x)| < 1.

209

## If be the root of the equation f (x) = 0 then

xn+1 = () (xn ) =  (0 )( xn )
where 0 lies between and xn .
Let xn1 , xn and xn+1 be three successive approximations to the root .
Then
xn = a( xn1 ) where a =  (0 )
xn+1 = a( xn ).
Eliminating a from these equations, we nd the relation
xn1
xn
=
xn+1
xn
which gives
= xn+1

(xn+1 xn )2
.
xn+1 2xn + xn1

(4.22)

xn = xn+1 xn ,

= xn+1

(xn )2
,
2 xn1

(4.23)

## which is known as Aitkens 2 -process.

Example 4.5.1 Find a root of the equation cos x xex = 0 using Aitkens 2 process.
Solution. Let x = ex cos x = (x) (say) and x0 = 0.
x1 = (x0 ) = 0.54030, x2 = (x1 ) = 0.49959.
x1 = x2 x1 = 0.04071, 2 x0 = x2 2x1 + x0 = 0.58101.
(x1 )2
(0.04071)2
= 0.50244.
=
0.49959

2 x0
0.58101
The results for n = 1, 2, 3, 4 are shown below.

Then x3 = x2

n
1
2
3
4

xn1
0.00000
0.54030
0.49958
0.50244

xn
0.54030
0.49959
0.50244
0.50225

xn+1
0.49959
0.50244
0.50225
0.50226

xn
0.04071
0.00285
0.00019
0.00001

2 xn1
0.58101
0.04356
0.00304
0.00021

xn+2
0.50244
0.50225
0.50226
0.50226

4.6

## Let x0 be an approximate root of the equation f (x) = 0. Suppose x1 = x0 + h be the

exact root of the equation, where h is the correction of the root (error). Then f (x1 ) = 0.
Using Taylors series, f (x1 ) = f (x0 + h) is expanded in the following form
f (x0 ) + hf  (x0 ) +

h2 
f (x0 ) + = 0.
2!

Neglecting the second and higher order derivatives the above equation reduces to
f (x0 ) + hf  (x0 ) = 0 or, h =

f (x0 )
.
f  (x0 )

Hence,
x1 = x0 + h = x0

f (x0 )
.
f  (x0 )

(4.24)

To compute the value of h, the second and higher powers of h are neglected so the
f (x0 )
is not exact, it is an approximate value. So, x1 , obtained from
value of h = 
f (x0 )
(4.24) is not a root of the equation, but it is a better approximation of x than x0 .
In general,
xn+1 = xn

f (xn )
.
f  (xn )

(4.25)

## This expression generates a sequence of approximate values x1 , x2 , . . . , xn , . . . each

successive term of which is closer to the exact value of the root than its predecessor.
The method will terminate when |xn+1 xn | becomes very small.
In Newton-Raphson method the arc of the curve y = f (x) is replaced by a tangent
to the curve, hence, this method is sometimes called the method of tangents.
Note 4.6.1 The Newton-Raphson method may also be used to nd a complex root of
an equation when the initial guess is taken as a complex number.
Geometrical interpretation
The geometrical interpretation of Newton-Raphson method is shown in Figure 4.7.
In this method, a tangent is drawn at (x0 , f (x0 )) to the curve y = f (x). The tangent
cuts the x-axis at (x1 , 0). Again, a tangent is drawn at (x1 , f (x1 )) and this tangent cuts
x-axis at (x2 , 0). This process is continued until xn = as n .

211

f (x)
6
6

6

6

x2

x1

x0

- x

## Figure 4.7: Geometrical interpretation of Newton-Raphson method.

The choice of initial guess of Newton-Raphson method is very important. If the initial
guess is near the root then the method converges very fast. If it is not so near the root
or if the starting point is wrong, then the method may lead to an endless cycle. This
is illustrated in Figure 4.8. In this gure, the initial guess x0 gives the fast convergence
to the root, the initial guess y0 leads to an endless cycle and the initial guess z0 gives a
divergent solution as f  (z0 ) is very small.
f (x)
6
6

z0
O

x0

y0

- x

## Figure 4.8: Illustration of the choice of initial guess in Newton-Raphson method.

Even if the initial guess is not close to the exact root, the method may diverges. To
choose the initial guess the following rule may be followed:
The endpoint of the interval [a, b] at which the sign of the function coincides with the
sign of the second derivative must be taken as the initial guess. When f (b) f  (x) > 0,
the initial guess is x0 = b, and when f (a) f  (x) > 0 then x0 = a be the initial guess.
Three dierent cases - divergent, cyclic and oscillation of Newton-Raphson method
are discussed in the following by examples.
Sometimes, if the initial guess x0 is far away from the exact root then the sequence

## 212 Numerical Analysis

{xn } may converges to some other root. This situation happens when the slope f  (x0 )
is small and the tangent to the curve y = f (x) is nearly horizontal. For example, if
f (x) = cos x and we try to nd the root = /2 starting with x0 = 3 then x0 =
4.01525, x2 = 4.85266, . . . and the sequence {xn } will converge to a dierent root
4.71239  3/2.
We consider another example which will produces a divergent sequence. Let
f (x) = xex

and

x0 = 2.

Then x1 = 4.0, x2 = 5.33333, . . . , x15 = 19.72255, . . . and clearly {xn } diverges slowly
to (Figure 4.9).
f (x)
6

U
6

- x
R6s

Figure 4.9: Newton-Raphson method produces a divergent sequence for f (x) = xex .
Now consider a function f (x) = x3 x 3 which will produce a cyclic sequence when
initial guess is x0 = 0. The sequence is
x1 = 3.0, x2 = 1.961538, x3 = 1.147176, x4 = 0.006579,
x5 = 3.000389, x6 = 1.961818, x7 = 1.147430, . . .
and it may be noted that xk+4  xk , k = 0, 1, 2, . . . (Figure 4.10).
But, the initial guess x0 = 2 gives the convergent sequence x1 = 1.72727, x2 =
1.67369, x3 = 1.67170, x4 = 1.67170.
The function f (x) = tan1 x and x0 = 1.45 gives a divergent oscillating sequence. If
the initial guess is x0 = 1.45 then
x1 = 1.55026, x2 = 1.84593, x3 = 2.88911, . . .
(Figure 4.11). But, if x0 = 1.5 then x1 = 0.07956, x2 = 0.00034, x3 = 0.00000.

x 

x1

x2

x3


?

x0

213

- x

y = x3 x 3
?

when x0 = 0.
4.6.1

## The Newton-Raphson iteration formula (4.25) is

xn+1 = xn

f (xn )
.
f  (xn )

Comparing this expression with xed point iteration formula xn+1 = (xn ) and we
obtain
f (xn )
.
(xn ) = xn 
f (xn )
This can be written as
(x) = x

f (x)
.
f  (x)

It is already proved that the iteration method converges if | (x)| < 1. Therefore,
Newton-Raphson method converges, if


d

f
(x)


(4.26)
or
|f (x) f  (x)| < |f  (x)|2
 dx x f  (x)  < 1
within the interval under consideration. Newton-Raphson method converges if the initial
guess x0 is chosen suciently close to the root and the functions f (x), f  (x) and f  (x)
are continuous and bounded in any small interval containing the root. The rate of
convergent of Newton-Raphson method is stated in the following theorem.

y
6

y = tan1 x
66

x  )
x3

x0

x1

x2

- x
6

?
?

## Figure 4.11: Newton-Raphson method produces a divergent oscillating sequence for

f (x) = tan1 x when x0 = 1.45.
Theorem 4.4 The rate of convergence of Newton-Raphson method is quadratic.
Proof. Let be a root of the equation f (x) = 0. Then f () = 0. The iteration scheme
for Newton-Raphson method is
xn+1 = xn

f (xn )
.
f  (xn )

Let xn = n + .
Therefore, above relation becomes
f (n + )
f  (n + )
f () + n f  () + (2n /2)f  () +
n
[by Taylors series]
f  () + n f  () +

2n f  ()

f () n + 2 f  () +

[as f () = 0]
n
f  ()

f () 1 + n f  () +

2n f  ()
f  ()
+ 1 n 
+
n n +
2 f  ()
f ()
f  ()
2 f  ()
+ 2n 
+ O(3n )
n 
2 f ()
f ()
1 2 f  ()

+ O(3n ).
2 n f  ()

n+1 + = n +
n+1 =

=
=
=

## Sol. of Algebraic and Transcendental Equs.

215

Neglecting the terms of order 3n and higher powers the above expression becomes
n+1 = A2n , where A =

f  ()
.
2f  ()

(4.27)

This relation shows that Newton-Raphson method has quadratic convergence or second order convergence.
Example 4.6.1 Use Newton-Raphson method to nd a root of the equation x3 +
x 1 = 0.
Solution. Let f (x) = x3 + x 1. Then f (0) = 1 < 0 and f (1) = 1 > 0. So one
root lies between 0 and 1. Let x0 = 0 be the initial root.
The iteration scheme is
f (xn )
f  (xn )
2x3 + 1
x3 + xn 1
= n2
.
= xn n 2
3xn + 1
3xn + 1

xn+1 = xn

n
0
1
2
3
4

xn
0
1
0.7500
0.6861
0.6823

xn+1
1
0.7500
0.6861
0.6823
0.6823

## Therefore, a root of the equation is 0.682 correct up to three decimal places.

Example 4.6.2 Find an iteration scheme to nd the kth root of a number a.
Solution. Let x be the kth root of a. That is, x = a1/k or xk a = 0.
Let f (x) = xk a. The iteration scheme is
f (xn )
f  (xn )
xk a
k xkn xkn + a
= xn n k1 =
k xn
k xk1
n

1
a 
=
(k 1)xn + k1 .
k
xn

xn+1 = xn
or, xn+1

## 216 Numerical Analysis

Example 4.6.3 Write down an iteration scheme for nding square root of a positive
number N . Hence nd the square root of the number 2.

## Solution. Let the square root of N be x. That is,x = N , or, x2 N = 0.

Thus the root of this equation is the required value of N . Let f (x) = x2 N . By
Newton-Raphson method
f (xn )
x2 N
1
N
.
= xn n
xn +
=
xn+1 = xn 
f (xn )
2xn
2
xn
This is the required iteration
scheme.

Second Part. Let x = 2 or, x2 2 = 0. Also, let f (x) = x2 2. Then f  (x) = 2x.
The Newton-Raphson iteration scheme is
x2 2
x2 + 2
= n
.
xn+1 = xn n
2 xn
2 xn
Let x0 = 1. The successive calculations are shown in the following.
n
0
1
2
3
Therefore, the value of

xn
1
1.50000
1.41667
1.41422

xn+1
1.5000
1.41667
1.41422
1.41421

## Example 4.6.4 Find a root of the equation x log10 x = 4.77 by Newton-Raphson

method correct up to ve decimal places.
Solution. Let f (x) = x log10 x 4.77. Here f (6) = 0.10109 < 0 and f (7) =
1.14569 > 0.
Therefore, one root lies between 6 and 7. Let the initial guess be x0 = 6.
The iteration scheme is
0.43429xn + 4.77
xn log10 xn 4.77
=
.
xn+1 = xn
log10 xn + log10 e
log10 (2.71828xn )
The values of x0 , x1 , x2 are shown in the following table.
n
xn
xn+1
0 6.000000 6.083358
1 6.083358 6.083152
2 6.083152 6.083153
Therefore, one root is 6.08315 correct up to ve decimal places.

## Sol. of Algebraic and Transcendental Equs.

217

3 x2n + 2
is an iteration scheme to
8
nd a root of the equation f (x) = 0. Find the function f (x).

## Solution. Let be the root obtained by performing the iteration scheme

xn+1 =

3 x2n + 2
.
8

Therefore, lim xn = .
n

1
Thus, lim xn+1 = 3 lim x2n + 2 .
n
8 n
1
This gives = [32 + 2], i.e., 32 + 2 = 8 or, 32 8 + 2 = 0.
8
Thus the required equation is 3x2 8x + 2 = 0 and hence f (x) = 3x2 8x + 2.
Example 4.6.6 Discuss the Newton-Raphson method to nd the root of the equation x10 1 = 0 starting with x0 = 0.5.
Solution. The real roots of this equation are 1.
Here f (x) = x10 1.
Therefore,
x10 1
9x10 + 1
xn+1 = xn n 9 = n 9 .
10xn
10xn
9 (0.5)10 + 1
= 51.65, which is far away from the root 1.
10 (0.5)9
This is because 0.5 was not close enough to the root x = 1.
But the sequence {xn } will converge to the root 1, although very slowly.
The initial root x0 = 0.9 gives the rst approximate root x1 = 1.068, which is close
to the root 1.
This example points out the role of initial approximation in Newton-Raphson method.
When x0 = 0.5 then x1 =

## Example 4.6.7 Find a complex root of the equation z 3 + 2z 2 + 2z + 1 = 0 starting

with the initial guess 0.5 + 0.5i.
Solution. Let z0 = 0.5 + 0.5i = (0.5, 0.5) be the initial guess and f (z) =
z 3 + 2z 2 + 2z + 1. Then f  (z) = 3z 2 + 4z + 2. The iteration scheme is
zn+1 = zn
All the calculations are shown below.

f (zn )
.
f  (zn )

## 218 Numerical Analysis

n
0
1
2
3
4
5
6

zn
(-0.50000, 0.50000)
(-1.00000, 1.00000)
(-0.75000, 0.75000)
(-0.41781, 0.71918)
(-0.55230, 0.85744)
(-0.49763, 0.86214)
(-0.50003, 0.86603)

f (zn )
( 0.25000, 0.25000)
( 1.00000,0.00000)
( 0.34375, 0.09375)
( 0.05444, 0.24110)
( 0.08475,-0.02512)
(-0.00014, 0.00785)
( 0.00004,-0.00003)

f  (zn )
(0.00000, 0.50000)
(-2.00000,-2.00000)
(-1.00000,-0.37500)
(-0.69919, 1.07384)
(-1.49971, 0.58836)
(-1.47746, 0.87439)
(-1.50005, 0.86587)

zn+1
(-1.00000, 1.00000)
(-0.75000, 0.75000)
(-0.41781, 0.71918)
(-0.55230, 0.85744)
(-0.49763, 0.86214)
(-0.50003, 0.86603)
(-0.50000, 0.86603)

4.7

## The Newton-Raphson method can be used to nd the multiple root of an equation.

But, its generalized form
xn+1 = xn p

f (xn )
f  (xn )

(4.28)

gives a faster convergent sequence. The term p1 f  (xn ) is the slope of the straight line
passing through (xn , f (xn )) and intersecting the x-axis at the point (xn+1 , 0). The
formula (4.28) reduces to Newton-Raphson formula when p = 1.
If is a root of f (x) = 0 with multiplicity p, then is also a root of f  (x) = 0
with multiplicity (p 1), of f  (x) = 0 with multiplicity (p 2) and so on. Hence the
expression
x0 p

f (x0 )
,
f  (x0 )

x0 (p 1)

f  (x0 )
,
f  (x0 )

x0 (p 2)

f  (x0 )
,...
f  (x0 )

should have the same value if there is a root with multiplicity p, when the initial guess
is very close to the exact root .
Theorem 4.5 The rate of convergence of the formula (4.28) is quadratic.
Proof. Let be a multiple root of multiplicity p, of the equation f (x) = 0. Then
f () = f  () = f  () = = f p1 () = 0 and f p () = 0. Let n = xn . Then from
(4.28),
n+1 = n p
= n p

f (n + )
f  (n + )
f () + n f  () + +
f  () + n f  () + +

p1
p1 ()
n
(p1)! f
p2
n

(p2)! f

p1 ()

pn p
p! f ()
p1
n

(p1)! f

p+1
p+1 ()
n
(p+1)! f

p ()

pn
p!

f p+1 () +

## Sol. of Algebraic and Transcendental Equs.

= n p

=
=
=
=

pn p
p! f ()
p1
n

p+1
p+1 ()
n
(p+1)! f

219

pn

f p+1 () +

1

2n
f p+1 ()
n f p+1 ()
n
+
+ 1 +
+
n p
p
p(p + 1) f p ()
p f p ()

2n
f p+1 ()
n f p+1 ()
n
+
+ 1
+
n p
p
p(p + 1) f p ()
p f p ()

2 f p+1 () 2n f p+1 ()
n n + n

+
p + 1 f p ()
p f p ()

f p+1 ()
1
2
+ O(3n ).
n
p(p + 1) f p ()
(p1)! f

p ()

p!

f p+1 ()
1
.
p(p + 1) f p ()
This shows that the rate of convergence is quadratic.
Thus n+1 = A2n , where A =

## Example 4.7.1 Find the double root of the equation x3 3x2 + 4 = 0.

Solution. Let x0 = 1.5 and f (x) = x3 3x2 + 4.
f  (x) = 3x2 6x, f  (x) = 6x 6.
0.625
f (x0 )
= 1.5 2
= 2.05556 and
x1 = x0 2 
f (x0 )
2.25
2.25
f  (x0 )
x1 = x0 2 
= 1.5
= 2.25000.
f (x0 )
3
The close values of x1 indicates that there is a double root near 2. Let x1 = 2.05556.
0.00943
f (x1 )
Then x2 = x1 2 
= 2.05556 2.
= 2.00051
f (x1 )
0.34262
0.34262
f  (x1 )
x2 = x1 2 
= 2.05556
= 2.00146.
f (x1 )
6.33336
Thus there is a double root at x = 2.00051 which is suciently close to the actual
root 2.
The Newton-Raphson method with same initial guess x0 = 1.5 produces the sequence
x1 = 1.77778, x2 = 1.89352, x3 = 1.94776, x4 = 1.97410, x5 = 1.98714,
x6 = 1.99353, x7 = 1.99689, x8 = 1.99850, x9 = 1.99961, x10 = 1.99980.
Thus at 10th iteration the Newton-Raphson method produces the root 2.000 correct
up to three decimal places, while the formula (4.28) needs only two iterations.

4.8

## Modication on Newton-Raphson Method

In the Newton-Raphson method, the derivative of the function f (x) is calculated at each
point xn . That is, at each iteration two functions are evaluated at xn , n = 0, 1, 2, . . ..

## 220 Numerical Analysis

But, some functions take much time to evaluate the derivative. To save this time one
can change the iteration scheme of Newton-Raphson method as
xn+1 = xn

f (xn )
.
f  (x0 )

(4.29)

That is, the derivative of f (x) is calculated only at the initial guess instead of several
dierent points xn . This method reduces the time for calculating the derivatives. But,
the rate of convergence of this method is linear, which is proved in Theorem 4.6.
Theorem 4.6 The rate of convergence of the formula (4.29) is linear.
Solution. Let be the root of the equation f (x) = 0. Then f () = 0, and n = xn .
Therefore from (4.29),
f (n + )
f () + n f  () +
n+1 = n
=

n
f  (x0 )
f  (x0 )


f  ()
+ O(2n ).
= n 1 
f (x0 )
Neglecting 2n and higher powers of 2n and denoting A = 1

f  ()
the above error
f  (x0 )

term becomes
n+1 = An .

(4.30)

This proved that the rate of convergence of the formula (4.29) is linear.
f (x)
6

6
s
6/

x3x2 x1

x0

## Sol. of Algebraic and Transcendental Equs.

221

Geometrical interpretation
The gradient of tangent at the point xn is f  (x0 ) for all n.
Thus the line passing through the point (xn , f (xn )) is parallel to the tangent drawn
at (x0 , f (x0 )), i.e., the tangent at (xn , f (xn )) in Newton-Raphson method is replaced
by a line parallel to the tangent drawn at (x0 , f (x0 )) and passing through the point
(xn , f (xn )). This phenomena is shown in Figure 4.12.
Example 4.8.1 Find a root of the equation x3 x + 1 = 0 using formula (4.29)
and Newton-Raphson method up to four decimal places.
Solution. One root of this equation lies between 2 and 1. Let x0 = 1.5 and
f (x) = x3 x + 1. Then f  (x0 ) = 5.75.
The iteration scheme of the formula (4.29) is
f (xn )
f  (x0 )
x3 xn + 1
1
=
(x3 6.75xn + 1).
= xn n
5.75
5.75 n

xn+1 = xn

## All the calculations are shown in the following table.

n
0
1
2
3
4
5
6
7

xn
1.50000
1.34783
1.33032
1.32614
1.32508
1.32481
1.32474
1.32472

xn+1
1.34783
1.33032
1.32614
1.32508
1.32481
1.32474
1.32472
1.32472

Therefore, one root of the given equation is 1.3247, correct up to four decimal places
attained at 7th iteration.
Using Newton-Raphson method
The iteration scheme for Newton-Raphson method is
f (xn )
f  (xn )
2x3 1
x3 xn + 1
= n2
.
= xn n 2
3xn 1
3xn 1

xn+1 = xn

## 222 Numerical Analysis

Let x0 = 1.5. The successive calculations are shown below.
n
0
1
2
3

xn
1.50000
1.34783
1.32520
1.32472

xn+1
1.34783
1.32520
1.32472
1.32472

Therefore, a root is 1.3247 correct up to four decimal places, attained at 3rd iteration.
This example shows that Newton-Raphson method is more faster than the method
given by (4.29).

4.9

xn+1 = xn

f  (xn

f (xn )
= (xn ) (say)
+ a(xn )f (xn ))

(4.31)

That is,
(x) = x

f  (x

f (x)
,
+ a(x)f (x))

## where a(x) is a smooth function.

Now,
f  (x)
 (x) = 1 
f (x + a(x)f (x))
f (x)f  (x + a(x)f (x))(1 + a (x)f (x) + a(x)f  (x))
+
{f  (x + a(x)f (x))}2

(4.32)

f  (x)
f  (x + a(x)f (x))
f  (x)f  (x + a(x)f (x))(1 + a (x)f (x) + a(x)f  (x))
+ 2
{f  (x + a(x)f (x))}2
f (x){f  (x + a(x)f (x))}2 {1 + a (x)f (x) + a(x)f  (x)}2
2
{f  (x + a(x)f (x))}3

and  (x) =

1
H.H.H.Homeier, A modied Newton method for root nding with cubic convergence, J. Computational and Applied Mathematics, 157 (2003) 227-230.

223

## f (x)f  (x + a(x)f (x)){1 + a (x)f (x) + a(x)f  (x)}2

{f  (x + a(x)f (x))}2
2

{f (x)} f (x + a(x)f (x))a (x)
+
{f  (x + a(x)f (x))}2
f (x)f  (x + a(x)f (x)){2a (x)f  (x) + a(x)f  (x)}
+
.
{f  (x + a(x)f (x))}2
+

## If be the root of the equation f (x) = 0 then f () = 0 and hence () = ,  () = 0

[from (4.32)] and
f  () 2f  ()f  (){1 + a()f  ()}
+
f  ()
{f  ()}2
f  ()
{1 + 2a()f  ()}.
= 
f ()

 () =

(4.33)

1
1
then  () = 0. This is the easiest way to put a(x) = 
.
2f  ()
2f (x)
Hence the iteration scheme for modied Newton-Raphson method is
Now, if a() =

xn+1 = xn

f (xn )
1
, where a(xn ) = 
.
f  (xn + a(xn )f (xn ))
2f (xn )

(4.34)

Also we have
() = ,  () = 0,  () = 0.

(4.35)

## The rate of convergence of this method is evaluated below:

Theorem 4.7 The rate of convergence of modied Newton-Raphson method is cubic.
Proof. Let be the root of the equation f (x) = 0. Also, let n = xn .
Therefore, xn+1 = (xn ) = (n + )
2
3
or, n+1 + = () + n  () + n  () + n  () +
2!
3!
3

## or, n+1 = n  () + O(4n ).

3!
[Using the facts () = ,  () = 0,  () = 0 from (4.35)]
Neglecting the terms 4n and higher powers of 4n the above equation reduces to
n+1 = A3n , where A =

1 
().
3!

(4.36)

## 224 Numerical Analysis

This method evaluates three functions f (x), x + a(x)f  (x) and f  (x + a(x)f (x)) at
each iteration.
Example 4.9.1 Find a root of the equation x3 3x2 +4 = 0 using modied NewtonRaphson method, starting with x0 = 1.5.
Solution. Let f (x) = x3 3x2 + 4. f  (x) = 3x2 6x,
1
1
=
.
a(x) = 
2
2f (x)
(6x 12x)
5x3 9x2 4
x3 3x2 + 4
=
.
Let g(x) = x + a(x)f (x) = x
6x2 12x
6x2 12x
f (xn )
.
Then the iteration scheme is xn+1 = xn 
f (g(xn ))
The calculations for each value of n is listed below.
n
0
1
2
3
4
5
6

xn
1.50000
1.85202
1.95215
1.98420
1.99468
1.99803
1.99916

f (xn )
0.62500
0.06245
0.00676
0.00074
0.00008
0.00001
0.000002

g(xn )
1.63889
1.89000
1.96421
1.98816
1.99601
1.99852
1.99937

f  (g(xn ))
1.77546
0.62370
0.21090
0.07062
0.02389
0.00887
0.00378

xn+1
1.85202
1.95215
1.98420
1.99468
1.99803
1.99916
1.99969

Therefor, a root of the given equation is 2.000, correct up to three decimal places,
and this value is attained at 6th iteration, while Newton-Raphson method takes 10
iterations (see Example 4.7.1).
Algorithm 4.4 (Newton-Raphson method). This algorithm nds a root of the
equation f (x) = 0 by Newton-Raphson method, when f (x), f  (x) and initial guess
x0 are supplied.
Algorithm Newton-Raphson
// f d(x) is the derivative of f (x) and is the error tolerance.//
Input function f (x), f d(x);
Set x1 = x0 ;
do
Set x0 = x1 ;
Compute x1 = x0 f (x0 )/f d(x0 );
while (|x1 x0 | > );
Print The root is, x1 ;
end Newton-Raphson

## Sol. of Algebraic and Transcendental Equs.

225

Program 4.4
.
/* Program Newton-Raphson
Program to find a root of the equation x*x*x-3x+1=0 by NewtonRaphson method. f(x) and its derivative fd(x) are to be supplied. */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
int k=0; /* counts number of iterations */
float x1,x0; /* x0 is the initial guess */
float eps=1e-5; /* error tolerance */
float f(float x);
float fd(float x);
printf("\nEnter the initial guess x0 ");
scanf("%f",&x0);
x1=x0;
do
{
k++;
x0=x1;
x1=x0-f(x0)/fd(x0);
}while(fabs(x1-x0)>eps);
printf("One root is %8.5f obtained at %d th iteration ",x1,k);
} /* main */
/* definition of the function f(x) */
float f(float x)
{
return(x*x*x-3*x+1);
}
/* definition of the function fd(x) */
float fd(float x)
{
return(3*x*x-3);
}
A sample of input/output:
Enter the initial guess x0 1.1
One root is 1.53209 obtained at 7 th iteration

4.10

Secant Method

## The main drawback of Newton-Raphson method is to determine the derivatives at

several points. In many cases, calculation of derivatives takes much time. In some cases
closed form expression for f  (x) is not available.
To remove this drawback, the derivative f  (x) is approximated by the backward
dierence
f (xi ) f (xi1 )
f  (xi ) 
xi xi1
where xi and xi1 are two approximations to the root but need not require the condition
f (xi ) f (xi1 ) < 0.
Then from the Newton-Raphson method
xi+1 = xi

f (xi )
f (xi )(xi xi1 )
= xi
.

f (xi )
f (xi ) f (xi1 )

(4.37)

This formula is same as the formula for Regula-falsi method and this formula needs
two initial guess x0 and x1 of the root.
Note 4.10.1 Regula-falsi method need an interval where the root belongs to, i.e., if
[x0 , x1 ] is the interval then f (x0 ) f (x1 ) < 0. But, secant method needs two nearest
values x0 and x1 of the exact root and not necessarily f (x0 ) f (x1 ) < 0.
Geometrical interpretation
A geometrical interpretation of secant method is illustrated in Figure 4.13.
A secant is drawn connecting f (xi1 ) and f (xi ). The point where it cuts the x-axis
is xi+1 . Another secant is drawn connecting f (xi ) and f (xi+1 ) to obtain xi+2 , and so
on.
4.10.1

xn+1 = xn

## (xn xn1 )f (xn )

.
f (xn ) f (xn1 )

(4.38)

Let be the exact root of the equation f (x) = 0 and the error at the nth iteration is
n = xn . Also f () = 0.

227

f (x)
6

xi

xi+1

xi+2

xi1

## Figure 4.13: Geometrical interpretation of secant method.

Then (4.38) becomes
(n n1 )f (n + )
f (n + ) f (n1 + )
(n n1 )[f () + n f  () + (2n /2)f  () + ]
= n
(n n1 )f  () + 12 (2n 2n1 )f  () +

1

1
2n f  ()
f  ()
+ 1 + (n + n1 ) 
+
= n n +
2 f  ()
2
f ()
f  ()
1
+ O(2n n1 + n 2n1 ).
= n n1 
2
f ()

n+1 = n

1 f  ()
. This is a non-linear dierence equation and
2 f  ()
1/p
= Apn . Then n = Apn1 . This gives n1 = n A1/p .

## Thus n+1 = cn n1 where c =

to solve it, let n+1

.
Therefore, Apn = cn n A1/p , i.e., pn = cA(1+1/p) n
Equating the power of n on both sides, obtain the equation for p
1/p

p=1+

1+1/p

1
p

or

1
p = (1 5).
2

.
Positive sign gives p = 1.618. Hence n+1 = A1.618
n
Thus the rate of convergence of secant method is 1.618, which is smaller than the
Newton-Raphson method. Thus this method converges at a slower rate. However, this
method evaluates function only once in each iteration, but Newton-Raphson method

## 228 Numerical Analysis

evaluates two functions f and f  in each iteration. In this context, the secant method
is more ecient as compared to Newton-Raphson method.
Example 4.10.1 Find a root of the equation x3 8x 4 = 0 using secant method.
Solution. Let f (x) = x3 8x 4 = 0. One root lies between 3 and 4. Let the initial
x0 f (x1 ) x1 f (x0 )
.
approximation be x0 = 3, x1 = 3.5. The formula for x2 is x2 =
f (x1 ) f (x0 )
The calculations are shown below:
x0
3.0000
3.5000
3.0421
3.0497

f (x0 )
1.0000
10.8750
0.1841
0.0333

x1
3.5000
3.0421
3.0497
3.0514

f (x1 )
10.8750
0.1841
0.0333
0.0005

x2
3.0421
3.0497
3.0514
3.0514

f (x2 )
0.1841
0.0333
0.0005
0.0005

## Therefore, a root is 3.051 correct up to four signicant gures.

Algorithm 4.5 (Secant method). This algorithm nds a root of the equation
f (x) = 0 by secant method when two initial guesses x0 and x1 are supplied.
Algorithm Secant
// The iteration terminates when |f (x1 ) f (x0 )| is very small (in this case slope of
the secant is very small) and |f (x2 )| < , is the error tolerance, is a very small
quantity, taken as 0.//
Input function f (x);
1. f x0 = f (x0 ); f x1 = f (x1 );
if |f x1 f x0 | < then
Print Slope too small, the method does not give correct root or fail
Stop;
endif;
x0 f x1 x1 f x0
;
Compute x2 =
f x1 f x0
Compute f2 = f (x2 );
if |f2 | < then
Print A root is, x2 ;
Stop;
endif;
Set f x0 = f x1 ; f x1 = f x2 ;
Set x0 = x1 ; x1 = x2 ;
Go to 1;
end Secant

## Sol. of Algebraic and Transcendental Equs.

229

Program 4.5
.
/* Program Secant
Program to find a root of the equation x*sin(x)-1=0 by
secant method. It is assumed that a root lies between x0 and x1.*/
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
float x0,x1,x2,fx0,fx1,fx2;
float eps=1e-5; /* error tolerance */
float delta=1e-5; /* slope */
float f(float x);
printf("\nEnter the values of x0 and x1 ");
scanf("%f %f",&x0,&x1);
fx0=f(x0);
fx1=f(x1);
if(fabs(fx1-fx0)<delta){
printf("Slope too small
the method does not give correct root or fail");
exit(0);
}
do
{
x2=(x0*fx1-x1*fx0)/(fx1-fx0);
fx2=f(x2);
if(fabs(fx2)<eps){
printf("One root is %8.5f ",x2);
exit(0);
}
fx0=fx1; fx1=fx2;
x0=x1;
x1=x2;
}while(fabs(fx2)>eps);
} /* main */
/* definition of function f(x), it may change accordingly */
float f(float x)
{
return(x*sin(x)-1);
}

## 230 Numerical Analysis

A sample of input/output:
Enter the values of x0 and x1 0 1
One root is 1.11416

4.11

Chebyshev Method

Let us consider the equation f (x) = 0. The function f (x) is expanded by Taylors series
in the neighbourhood of xn as 0 = f (x) = f (xn ) + (x xn )f  (xn ) + .
f (xn )
.
This relation gives
x = xn 
f (xn )
This is the (n + 1)th approximation to the root. Therefore,
xn+1 = xn

f (xn )
.
f  (xn )

(4.39)

Again, expanding f (x) by Taylors series and retaining up to second order term,
shown below.
0 = f (x) = f (xn ) + (x xn )f  (xn ) +

(x xn )2 
f (xn )
2

(xn+1 xn )2 
f (xn ) = 0.
2
Substituting the value of xn+1 xn from (4.39) to the last term and we nd

Therefore,

## f (xn ) + (xn+1 xn )f  (xn ) +

Thus,

xn+1 = xn

1 [f (xn )]2 
f (xn ) = 0.
2 [f  (xn )]2

1 [f (xn )]2 
f (xn )

f (xn ).
f  (xn ) 2 [f  (xn )]3

(4.40)

## This formula is the extended form of Newton-Raphson formula and it is known as

Chebyshevs formula.
The rate of convergence of this method is cubic.

4.12

Muller Method

The main idea of this method is, the function f (x) is approximated by a quadratic
polynomial passing through the three points in the neighbour of the root. The root of
this quadratic is assumed to approximate the root of the equation f (x) = 0.
Let xn2 , xn1 , xn be any three distinct approximation to a root of the equation
f (x) = 0. We denote f (xn2 ) = fn2 , f (xn1 ) = fn1 and f (xn ) = fn .

231

## Let the quadratic polynomial be

f (x) = ax2 + bx + c.

(4.41)

Suppose, (4.41) passes through the points (xn2 , fn2 ), (xn1 , fn1 ) and (xn , fn ),
then
ax2n2 + bxn2 + c = fn2

(4.42)

ax2n1 + bxn1 + c =
ax2n + bxn + c = fn

(4.43)

fn1

(4.44)

## Eliminating a, b, c from (4.41)-(4.44), we obtain the following determinant



2

 f (x)
x
1
x


2
 fn2
xn2
xn2
1 

= 0.
 fn1
x2n1
xn1
1 

 fn
x2n
xn
1 
By expanding this determinant the function f (x) can be written as
(x xn2 )(x xn )
(x xn1 )(x xn )
fn2 +
fn1
(xn2 xn1 )(xn2 xn )
(xn1 xn2 )(xn1 xn )
(x xn2 )(x xn1 )
+
(4.45)
fn .
(xn xn2 )(xn xn1 )

f (x) =

## This is a quadratic polynomial passing through the given points.

Let h = x xn , hn = xn xn1 and hn1 = xn1 xn2 . Then above equation
reduces to
h(h + hn + hn1 )
h(h + hn )
fn2
fn1
hn1 (hn1 + hn )
hn hn1
(h + hn )(h + hn + hn1 )
+
fn = 0,
hn (hn + hn1 )

(4.46)

since f (x) = 0.
Now, introducing
=

h
,
hn

n =

hn
hn1

and

n = 1 + n .

## Therefore, the equation (4.46) reduces to the following form

2 (fn2 2n fn1 n n + fn n )n1
+ {fn2 2n fn2 n2 + fn (n + n )}n1 + fn = 0

(4.47)

## 232 Numerical Analysis

or,
2 cn + gn + n fn = 0,

(4.48)

## gn = 2n fn2 n2 fn1 + (n + n )fn

where

cn = n (n fn2 n fn1 + fn ).
The equation (4.48) now becomes
 
1
gn
+ cn = 0.
n fn 2 +

=

2 f
 n n
.
gn gn2 4n fn cn

## The sign in the denominator of (4.49) is according as gn > 0 or gn < 0.

Thus
x xn
=
or
x = xn + (xn xn1 ).
xn xn1

(4.49)

(4.50)

Now, replacing x on left hand side by xn+1 and obtain the formula
xn+1 = xn + (xn xn1 ),

(4.51)

## which is called the Muller method.

This method is also an iterative method and free from evaluation of derivative as in
Newton-Raphson method.
Example 4.12.1 Find a root of the equation x3 3x 5 = 0 using Mullers method
which lies between 2 and 3.
Solution. Let x0 = 2.0, x1 = 2.5 and x2 = 3.0, f (x) = x3 3x 5.
hn
, n = 1 + n
hn1
gn = 2n fn2 n2 fn1 + (n + n )fn
cn = n (n fn2 n fn1 + fn )
2 f
 n n
=
gn (gn2 4n fn cn )
xn+1 = xn + hn .

hn = xn xn1 , n =

233

n
2
3
4
5

xn2
2.00000
2.50000
3.00000
2.28252

xn1
2.50000
3.00000
2.28252
2.27897

xn
3.00000
2.28252
2.27897
2.27902

gn
23.50000
3.89279
-0.04477
0.00056

cn
3.75000
-1.74262
0.00010
0.00000

1.43497
0.00494
-0.01263
-0.00338

xn+1
2.28252
2.27897
2.27902
2.27902

## Therefore, one root is 2.2790 correct up to four decimal places.

Summary of the Methods
Method

Formula

1. Bisection

xn+1 =

Order of Evaluation
Convergent of function
in each step

xn + xn1
2
xn1 fn xn fn1
fn fn1

2. False Position

xn+1 =

3. Iteration

xn+1 = (xn )

4. Newton-Raphson xn+1

5. Secant

Gain of one

f (xn )
= xn 
f (xn )

xn+1 =

xn1 fn xn fn1
fn fn1

bit per
iteration
1

1.62

1.84

6. Modied
Newton-Raphson

xn+1 = xn

7. Chebyshev

xn+1 = xn

8. Muller

xn+1

f  (xn

fn
12 fn /fn )

fn 1 fn2 

f
fn
2 f  3n n
= xn + (xn xn1 )

## Example 4.12.2 Consider the iteration method xn+1 = (xn ), n = 0, 1, 2, . . . for

solving the equation f (x) = 0. If the iteration function is in the form
(x) = x f (x) {f (x)}2 {f (x)}3
where , and are arbitrary parameters then nd the values of , , such that
the iteration method has (i) third and (ii) fourth order convergence.

## 234 Numerical Analysis

Solution. Let be an exact root of the equation f (x) = 0, i.e., f () = 0.
Now,
 (x) = 1 f  (x) 2f (x)f  (x) 3{f (x)}2 f  (x)
 (x) = f  (x) 2{f  (x)}2 2f (x)f  (x) 6f (x){f  (x)}2
3{f (x)}2 f  (x)
 (x) = f  (x) 6f  (x)f  (x) 2f (x)f  (x) 6{f  (x)}3
18f (x)f  (x)f  (x) 3{f (x)}2 f  (x).
Substituting f () = 0 to the above equations.
() =
 () = 1 f  ()
 () = f  () 2{f  ()}2
 () = f  () 6f  ()f  () 6{f  ()}3 .
Let n = xn .
3
4
Then n+1 + = (n + ) = () + n  () + 12 2n  () + 6n  () + 24n iv () +
(i) For third order convergence, the value of  () and  () should be zero. In this
case,
3
or,
n+1 = A3n .
n+1 = n  () +
6
Thus, 1 f  () = 0 and f  () 2{f  ()}2 = 0.
f  ()
1
=
.
That is, =  ,
f ()
2{f  ()}3
Hence, for third order convergence, the values of and are given by
=

1
,

f (x)

f  (x)
.
2{f  (x)}3

## (ii) For the fourth order convergence,  () =  () =  () = 0.

In this case n+1 = A4n .
f  (x)
1
,
=
Then = 
f (x)
2{f  (x)}3
and f  () 6f  ()f  () 6{f  ()}3 = 0.
That is,
6{f  (x)}3 = f  (x) 6f  (x)f  (x) =
or,

{f  (x)}2
f  (x)
+
.
6{f  (x)}4 2{f  (x)}5

f  (x)
f  (x)
f  (x)f  (x)
+
6
f  (x)
2{f  (x)}3

235

## Example 4.12.3 The equation x3 5x2 + 4x 3 = 0 has one root near x = 4,

which is to be computed by the iteration
x0 = 4
xn+1 =

## 3 + (k 4)xn + 5x2n x3n

,
k

k is integer.

Find the value of k such that the iteration scheme has fast convergence.
Solution. Let lim xn = lim xn+1 = .
n
n
Since is a root of the given equation,
3 5 2 + 4 3 = 0.
Substituting, xn = + n , xn+1 = + n+1 to the given iteration scheme. Then
k(n+1 + ) = 3 + (k 4)(n + ) + 5(n + )2 (n + )3
kn+1 = (3 4 + 5 2 3 ) + n {(k 4) + 10 3 2 }
+2n (5 3) 3n
or, kn+1 = n {(k 4) + 10 3 2 } + 2n (5 3) 3n .
If k 4 + 10 3 2 = 0 or, k = 3 2 10 + 4 then
n+1 = A2n , where A =

1
(5 3)
k

## i.e., the rate of convergence is 2.

The values of is determined from the equation
3 5 2 + 4 3 = 0.
The Newton-Raphson method is used to determine the approximate value of as
= 4.22069. Thus the required value of k is k = 3 2 10 + 4 = 15.23577.
Example 4.12.4 Consider the following iteration scheme
xn+1 = xn

af (xn )
f  (xn bf (xn )/f  (xn ))

where a, b are arbitrary parameters, for solving the equation f (x) = 0. Determine
a and b such that the iteration method is of order as high as possible for nding a
simple root of f (x) = 0.

## 236 Numerical Analysis

Solution. Here the iteration scheme is xn+1 = (xn ) where
(x) = x

af (x)
af (x)
=x 
f  (x bf (x)/f  (x))
f (g(x))

where
g(x) = x

bf (x)
,
f  (x)

g  (x) = 1

.
{f  (x)}2

## If be a root of f (x) = 0 then f () = 0, g() = , g  () = 1 b.

af  (x)f  (g(x)) af (x)f  (g(x))g  (x)
{f  (g(x))}2


af ()f ()
 () = 1
= 1 a.
{f  ()}2

 (x) = 1

## Now, we choose 1 a = 0 so that  () = 0.

Then
f (x)f  (g(x))g  (x)
f  (x)
+
f  (g(x))
{f  (g(x))}2


f (x)f (g(x)) f  (x)f  (g(x))g  (x)
 (x) =
{f  (g(x))}2
f  (x)f  (g(x))g  (x)
+
{f  (g(x))}2
{f (x)f  (g(x)){g  (x)}2 + f (x)f  (g(x))g  (x)}{f  (g(x))}2
+
{f  (g(x))}4



f (x)f (g(x))g (x) 2f (g(x)) f  (g(x))g  (x)

{f  (g(x))}4
f  ()f  () f  ()f  ()(1 b)
 () =
{f  ()}2


f ()f ()(1 b)
f  ()
{1 + 2(1 b)}.
+
=
{f  ()}2
f  ()
 (x) = 1

## Now,  () = 0 if 1 + 2(1 b) = 0 or, b = 1/2.

From the relation, xn+1 = (xn ) one can write
n+1 + = (n + ) = () + n  () +
or, n+1 = n  () +

2n 
3
() + n  () +
2
6

2n 
3
() + n  () +
2
6

237

## If a = 1 and b = 1/2 then  () = 0 and  () = 0. Then

n+1 

3n 
().
6

Hence the iteration scheme will have a third order convergence when a = 1 and
b = 1/2.

4.13

## The polynomial of degree n is generally denoted by Pn (x) and is dened as

Pn (x) a0 xn + a1 xn1 + + an1 x + an = 0

(4.52)

## where a0 , a1 , . . . , an are real coecients.

A number is a root of the polynomial Pn (x) i Pn (x) is exactly divisible by x .
If Pn (x) is exactly divisible by (x )k (k 1), but is not divisible by (x )k+1 ,
then is a k-fold root or a root of multiplicity k of the polynomial Pn (x). The roots of
multiplicity k = 1 are called simple root or single root.
The following theorem assures the existence of the roots of a polynomial equation.
Theorem 4.8 (The fundamental theorem of algebra). Every polynomial equation
with any numerical coecients whose degree is not lower than unity has at least one
root, real or complex.
From this theorem it can be proved that:
Every polynomial Pn (x) of degree n (n 1) with any numerical coecients has exactly
n roots, real or complex.
The roots of the equation (4.52) may be real or complex. If the coecients of (4.52)
are real and has a complex root + i of multiplicity k then (4.52) has a complex root
i also of multiplicity k.
The number of positive and negative roots of a polynomial equation can be determined
by using Descartes rule of signs:
The number of positive real roots of the algebraic equation Pn (x) = 0 with real coecients
either is equal to the number of sign changes in the sequence of the coecients of the
equation Pn (x) = 0 or is less than the number of sign changes by an even integer. The
number of negative roots of the equation is equal to the number of sign changes in the
sequence of coecients of Pn (x) or is smaller by an even integer.
The Sturms theorem gives the exact number of positive real roots of a polynomial
equation.
Let f (x) be a given polynomial of degree n and let f1 (x) be its rst order derivative.
Let f2 (x) be the remainder of f (x) when it is divided by f1 (x) taken with reverse sign.

## 238 Numerical Analysis

Similarly, f3 (x) is the remainder of f1 (x) when it is divided by f2 (x) with the reverse sign
and so on. This division process is terminated when the quotient becomes a constant.
Thus we obtain a sequence of function f (x), f1 (x), f2 (x), . . . , fn (x) called the Sturm
functions or the Sturm sequences.
Theorem 4.9 (Sturm). The number of real roots of the equation f (x) = 0 on [a, b] is
equal to the dierence between the number of changes of sign in the Sturm sequence at
x = a and x = b, provided f (a) = 0 and f (b) = 0.
4.13.1

Domains of roots

## Let the polynomial of degree n be

a0 xn + a1 xn1 + + an1 x + an = 0,

(4.53)

where a0 , a1 , . . . , an are real coecients, and let A = max{|a1 |, |a2 |, . . . , |an |} and B =
max{|a0 |, |a1 |, . . . , |an1 |}. Then the roots of the equation (4.53) lie in the interval
r < |x| < R where
r=

1
1 + B/|an |

and

R=1+

A
.
|a0 |

(4.54)

Here r is the lower bound and R is the upper bound of the positive roots of the
equation (4.53) and R and r are the lower and the upper bounds of the negative
roots respectively.
The Lagranges or Newtons method may also be used to nd the upper bound of the
positive roots of the equation (4.53).
Theorem 4.10 (Lagranges). If the coecients of the polynomial
a0 xn + a1 xn1 + + an1 x + an = 0
satisfy the conditions a0 > 0, a1 , a2 , . . . , am1 0, am <
 0, for some m n, then the
upper bound of the positive roots of the equation is 1 + m B/a0 , where B is the greatest
of the absolute values of the negative coecients of the polynomial.
Theorem 4.11 (Newtons). If for x = c the polynomial
f (x) a0 xn + a1 xn1 + + an1 x + an = 0
and its derivatives f  (x), f  (x), . . . assume positive values then c is the upper bound of
the positive roots of the equation.
The roots of a polynomial equation can be determined in two techniques iteration
methods and direct methods. In this section, two iteration methods, viz., Birge-Vieta
and Bairstow methods, and one direct method Graees root squaring method are
discussed.

## Sol. of Algebraic and Transcendental Equs.

239

Iterative Methods
4.14

Birge-Vieta Method

This method is based on the Newton-Raphson method. Here a real number is determined such that (x ) is a factor of the polynomial
Pn (x) = xn + a1 xn1 + a2 xn2 + + an1 x + an = 0.

(4.55)

Let Qn1 (x) and R be the quotient and remainder when Pn (x) is divided by the
factor (x ), where Qn1 (x) is a polynomial of degree (n 1) of the form
Qn1 (x) = xn1 + b1 xn2 + b2 xn3 + + bn2 x + bn1 .

(4.56)

(4.57)

Thus

## The value of R depends on . Now, the problem is to nd the value of starting

from an initial guess x0 such that R() = 0 or it is equivalent to
R() = Pn () = 0.

(4.58)

## The value of can be determined by Newton-Raphson method or any other method.

The Newton-Raphson methods for (4.58) is
xk+1 = xk

Pn (xk )
, k = 0, 1, 2, . . . .
Pn (xk )

(4.59)

The values of Pn (xk ) and Pn (xk ) can be determined by synthetic division. To
determine the values of b1 , b2 , . . . , bn1 and R, comparing the coecient of like powers
of x on both sides of (4.57) and obtain the following relations.
a1 = b1
a2 = b2 b1
..
.

b1 = a1 +
b2 = a2 + b1
..
.

ak = bk bk1
..
.

bk = ak + bk1
..
.

an = R bn1

R = an + bn1

From (4.57),
Pn () = R = bn (say).

(4.60)

## 240 Numerical Analysis

Hence,
bk = ak + bk1 , k = 1, 2, . . . , n, with b0 = 1.

(4.61)

## Thus bn is the value of Pn .

To determine the value of Pn , dierentiating (4.57) with respect to x
Pn (x) = (x )Qn1 (x) + Qn1 (x)
That is,
Pn () = Qn1 () = n1 + b1 n2 + + bn2 + bn1 .

(4.62)

+ + bn2 xi + bn1 .
Pn (xi ) = xin1 + b1 xn2
i

(4.63)

Thus

Pn (x) can be evaluated as Pn (x) is evaluated. Dierentiating (4.61) with respect to
and obtain
dbk1
dbk
= bk1 +
.
d
d
Let
dbk
= ck1 .
d

(4.64)

## Then the above relation becomes

ck1 = bk1 + ck2
This gives
ck = bk + ck1 , k = 1, 2, . . . , n 1.

(4.65)

Pn () =

dbn
dR
=
= cn1
d
d

[using (4.64)].

## Thus Newton-Raphson method becomes

xk+1 = xk

bn
cn1

, k = 0, 1, 2, . . . .

## This method is known as Birge-Vieta method.

The Table 4.2 is useful to determine bk and ck for hand calculations.

(4.66)

241

## Table 4.2: Scheme to calculate bs and cs.

x0 1 a1
x0
x0 1 b1
x0
1 c1

a2
x0 b1
b2
x0 c1
c2

an2
x0 bn3
bn2
x0 cn3
cn2

an1
an
x0 bn2
x0 bn1
bn1
bn = R
x0 cn2
cn1 = Pn (x0 )

Example 4.14.1 Find all the roots of the polynomial equation x4 8x3 +14.91x2 +
9.54x 25.92 = 0. One root of the equation lies between 1 and 2.
Solution. Let the polynomial be denoted by P4 (x). Also, let the initial guess be
x0 = 1.2.
1.2 1

8 14.91 9.54
25.92
1.2 8.16 8.10
21.168
1.2 1 6.8 6.75 17.64
4.752=b4 = P4 (x0 )
1.2 6.72 0.036
1 5.6 0.03 17.676=c3 = P4 (x0 )
Therefore,
x1 = x0

b4
4.752
= 1.46884.
= 1.2
c3
17.676

1.46884 1

8
14.91 9.54
25.92
1.46884 9.59323 7.80949
25.48362
1.46884 1 6.53116 5.31677 17.34949
0.43638=b4
1.46884 7.43574 3.11243
1 5.06232 2.11897 14.23706=c3
Then x2 = x1

b4
0.43638
= 1.49949.
= 1.46884
c3
14.23706

1.49949 1

8
14.91 9.54
25.92
1.49949 9.74745 7.74119
25.91298
1.49949 1 6.50051 5.16255 17.28119
0.00702=b4
1.49949 7.49898 3.50345
1 5.00102 2.33643 13.77774=c3
Then x3 = x2

b4
0.00702
= 1.50000.
= 1.49949
c3
13.77774

## 242 Numerical Analysis

Therefore, one root is 1.50000, which is an exact root. The reduce polynomial is
x3 6.50051x2 + 5.16255x + 17.28119 = 0.
One root of this equation lies between 4 and 5. Let x0 = 4.0.
4 1 6.50051
4
4 1 2.50051
4
1 1.49949
Therefore,
x1 = x0

5.16255
17.28119
10.00204 19.35796
4.83949
2.07677=b3
5.99796
1.15847= c2

b3
2.07677
= 5.79268.
=4
c2
1.15847

5.79268 1 6.50051
5.79268
5.79268 1 0.70783
5.79268
1 5.08485
x2 = x1

b3
23.43485
= 5.02476.
= 5.79268
c2
30.51723

5.02476 1 6.50051
5.02476
5.02476 1 1.47575
5.02476
1 3.54901
x3 = x2

5.16255
17.28119
7.41529
11.31948
2.25274
5.96171=b3
17.83292
15.58018= c2

b3
5.96171
= 4.64211.
= 5.02476
c2
15.58018

4.64211 1 6.50051
4.64211
4.64211 1 1.85840
4.64211
1 2.78371
x4 = x3

5.16255
17.28119
4.10023
6.15366
1.06232
23.43485=b3
29.45491
30.51723= c2

5.16255
17.28119
8.62690
16.08188
3.46435
1.19931=b3
12.92229
9.45794= c2

b3
1.19931
= 4.51531.
= 4.64211
c2
9.45794

243

## We take x = 4.5 as another root. The next reduced equation is

x2 1.85840x 3.46435 = 0
and roots of this equation are
x = 3.00953, 1.15113.
Hence the roots of the given equation are
1.5, 4.5, 3.00953, 1.15113.
But, the roots 3.00953 and 1.15113 contain some errors, because we approximate
4.51531 as 4.5. Some more iterations are required to achieve the root 4.5 and the
correct quotient polynomial.
Algorithm 4.6 (Birge-Vieta method). This algorithm nds a root of a polynomial equation when the coecients a1 , a2 , . . . , an and an initial guess x0 are given.
Algorithm Birge-Vieta
//n is the degree of the polynomial and N is the maximum number of iterations to
be performed. Assume that the leading coecient is one. is the error tolerance.//
Read x0 , , n, N ;
Read ai , i = 1, 2, . . . , n;
for i = 1 to N do
Set b0 = 1, c0 = 1;
for k = 1 to n do
Compute bk = ak + x0 bk1 ;
endfor;
for k = 1 to n 1 do
Compute ck = bk + x0 ck1 ;
endfor;
bn
;
Compute x1 = x0
cn1
if |x1 x0 | < then
PrintOne root is, x1 ;
Stop;
else
Set x0 = x1 ;
endif;
endfor;
end Birge-Vieta

## 244 Numerical Analysis

Program 4.6
.
/* Program Birge-Vieta for polynomial equation
Program to find a root of the polynomial equation by
Birge-Vieta method. Leading coefficient is 1 */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
int n, N,i,k;
float x0,x1,a,b,c;
float epp=1e-5; /* error tolerance */
printf("\nEnter the degree of the polynomial ");
scanf("%d",&n);
printf("Enter the coefficients of the polynomial,
for(i=1;i<=n;i++) scanf("%f",&a[i]);
printf("Enter the initial guess x0 ");
scanf("%f",&x0);
printf("Enter maximum number of iterations to be done ");
scanf("%d",&N);
b=1; c=1;
for(i=1;i<=N;i++)
{
for(k=1;k<=n;k++) b[k]=a[k]+x0*b[k-1];
for(k=1;k<=n-1;k++) c[k]=b[k]+x0*c[k-1];
x1=x0-b[n]/c[n-1];
if(fabs(x1-x0)<epp)
{
printf("One root is %8.5f obtained at %d iterations",x1,i);
printf("\nCoefficients of the reduced polynomial are\n ");
for(k=0;k<=n-1;k++) printf("%f ",b[k]);
exit(0);
}
else
x0=x1;
} /* i loop */
} /* main */

## Sol. of Algebraic and Transcendental Equs.

245

A sample of input/output:
Enter the degree of the polynomial 4
Enter the coefficients of the polynomial, except leading coeff.
-3 0 6 -4
Enter the initial guess x0 1
Enter maximum number of iterations to be done 100
One root is 1.00000 obtained at 1 iterations
Coefficients of the reduced polynomial are
1.000000 -2.000000 -2.000000 4.000000

4.15

Bairstow Method

## The roots of a polynomial equation can also be determined by extracting a quadratic

factor from the polynomial. The roots (real or complex) of a quadratic equation can
determine using a known closed form formula. The Bairstow method is used to
Let the polynomial Pn (x) of degree n be
xn + a1 xn1 + a2 xn2 + + an1 x + an = 0.

(4.67)

## Let x2 + px + q be a factor of (4.67). If (4.67) is divided by the factor x2 + px + q then

we obtain a polynomial Qn2 (x) of degree (n 2) and a remainder Rx + S of degree
one, where R and S are independent of x. Thus the polynomial Pn (x) can be written
as
Pn (x) = (x2 + px + q)Qn2 (x) + Rx + S
(4.68)
where

(4.69)

## The values of R and S depends on p and q. If x2 + px + q is a factor of Pn (x) then

R and S should be zero. Thus our problem is to determine the values of p and q such
that
R(p, q) = 0
and
S(p, q) = 0.
(4.70)
These equations are two non-linear equations in p and q. The values of p and q can
then be determined by Newton-Raphson method for two variables (see Section 4.17.3).
Let (pt , qt ) be the true values of p and q and p, q be the corrections to p and q.
Then
and
qt = q + q.
pt = p + p
Thus
R(pt , qt ) = R(p + p, q + q) = 0

and

S(pt , qt ) = S(p + p, q + q) = 0.

## 246 Numerical Analysis

Therefore, by Taylors series
R
R
+ q
+ = 0
p
q
S
S
+ q
+ = 0.
and S(p + p, q + q) = S(p, q) + p
p
q
R(p + p, q + q) = R(p, q) + p

The derivatives are evaluated at (p, q). Neglecting the square and higher powers of
p and q the above equations reduce to
pRp + qRq = R

(4.71)

pSp + qSq = S.

(4.72)

## The values of p and q are thus given by

p =

SRp RSp
RSq SRq
, q =
.
Rp Sq Rq Sp
Rp Sq Rq Sp

(4.73)

Now, we determine the coecient of the polynomial Qn2 (x) and the expression for
R and S in terms of p and q.
From equations (4.67)-(4.69)
xn + a1 xn1 + a2 xn2 + + an1 x + an
= (x2 + px + q)(xn2 + b1 xn3 + + bn3 x + bn2 ).

(4.74)

## Equating the coecients of xn , xn1 , . . . on both sides, we get

a1 = b1 + p
a2 = b2 + pb1 + q
..
.

b1 = a1 p
b2 = a2 pb1 q
..
.

ak = bk + pbk1 + qbk2
..
.

bk = ak pbk1 qbk2
..
.

an = S + qbn2

## R = an1 pbn2 qbn3

S = an qbn2 .

(4.75)

In general,
bk = ak pbk1 qbk2 ,

k = 1, 2, . . . , n

(4.76)

where b0 = 1, b1 = 0.
In this notation,
R = bn1 ,

S = bn + pbn1 .

(4.77)

## Sol. of Algebraic and Transcendental Equs.

247

Thus R and S are available when bs are known. To determine the partial derivatives
Rp , Rq , Sp and Sq , dierentiating (4.76) with respect to p and q.
bk
bk1
bk2
= bk1 p
q
,
p
p
p
bk2
bk1
bk
= bk2 p
q
,
q
q
q
Denoting

bk
= ck1 ,
p
bk
= ck2 .
and
q

## Then (4.78) becomes

and (4.79) becomes

b0
b1
=
=0
p
p
b0
b1
=
=0
q
q

k = 1, 2, . . . , n

(4.78)
(4.79)

(4.80)
(4.81)

(4.82)

(4.83)

## Thus the recurrence relation to determine ck using bk is

ck = bk pck1 qck2 , k = 1, 2, . . . , n 1 and c0 = 1, c1 = 0.
Therefore,

(4.84)

bn1
= cn2
p
bn
bn1
Sp =
+p
+ bn1 = bn1 cn1 pcn2
p
p
bn1
= cn3
Rq =
q
bn1
bn
+p
= (cn2 + pcn3 ).
Sq =
q
q

Rp =

To nd the explicit expression for p and q, substituting the above values in (4.73).
Therefore,
bn cn3 bn1 cn2
cn3 (cn1 bn1 )
bn1 (cn1 bn1 ) bn cn2
.
q = 2
cn2 cn3 (cn1 bn1 )

p =

c2n2

(4.85)

## Therefore, the improved values of p and q are p + p and q + q. Thus if p0 , q0 be

the initial values of p and q then the improved values are
p1 = p0 + p

and

q1 = q0 + q.

(4.86)

## 248 Numerical Analysis

The values of bk s and ck s may be calculated using the following scheme: (when p0
and q0 are taken as initial values of p and q).
1

a1
p0

b1
p0

c1

p0
q0
p0
q0

a2
p0 b1
q0
b2
p0 c1
q0
c2

ak
p0 bk1
q0 bk2
bk
p0 ck1
q0 ck2
ck

an1
p0 bn2
q0 bn3
bn1
p0 cn2
q0 cn3
cn1

an
p0 bn1
q0 bn2
bn

Once p1 and q1 are evaluated, the next improved values p2 , q2 are determined from
the relation
p2 = p1 + p,
q2 = q1 + q.
In general,

pk+1 = pk + p,

qk+1 = qk + q,

(4.87)

## the values of p and q are determined at p = pk and q = qk .

The repetition is to be terminated when p and q have been obtained to the desired
accuracy.
The polynomial
Qn2 (x) = Pn (x)/(x2 + px + q) = xn2 + b1 xn3 + + bn3 x + bn2
is called the deated polynomial. The next quadratic polynomial can be obtained in
similar process from the deated polynomial.
The rate of convergence of this method is quadratic as the computations of p and
q are based on Newton-Raphson method.
Example 4.15.1 Extract a quadratic factor using the Bairstow method from the
equation
x4 + 4x3 7x2 22x + 24 = 0.
Solution. Let the initial guess of p and q be p0 = 0.5 and q0 = 0.5. Then
1.00000

4.00000
0.50000

1.00000

3.50000 = b1
0.50000

1.00000

3.00000
= c1

0.5
0.5
0.5
0.5

7.00000
1.75000
0.50000
9.25000
1.50000
0.50000
11.25000
= c2

22.00000
4.62500
1.75000
19.12500
5.62500
1.50000
15.00000
= c3

24.00000
9.56250
4.62500
38.18750 = b4

## Sol. of Algebraic and Transcendental Equs.

p =

c22

b4 c1 b3 c2
= 0.88095,
c1 (c3 b3 )

q =

b3 (c3 b3 ) b4 c2
= 3.07143
c22 c1 (c3 b3 )

## Therefore, p1 = p0 + p = 1.38095, q1 = q0 + q = 2.57143.

Second iteration
1.00000

4.00000
1.38095

1.00000

2.61905
1.38095

1.00000

1.23810

1.38095
2.57143
1.38095
2.57143

7.00000
3.61678
2.57143
8.04535
1.70975
2.57143
7.18367

22.00000
11.11025
6.73469
4.15506
9.92031
3.18367
8.94893

24.00000
5.73794
20.68805
9.04989

p = 0.52695, q = 0.29857.
p2 = p1 + p = 1.90790, q2 = q1 + q = 2.86999.
Third iteration
1.00000
1.90790
2.86999
1.00000
1.90790
2.86999
1.00000

4.00000
1.90790
2.86999
2.09210
1.90790
0.18420

7.00000
3.99152
6.00432
8.12152
0.35144
2.86999
5.60297

22.00000
15.49504
23.30873
0.50064
10.68990
0.52866
10.71793

24.00000
0.95517
1.64644

p = 0.08531, q = 0.12304.
p3 = p2 + p = 1.99321, q3 = q2 + q = 2.99304.
Fourth iteration
1.00000

4.00000
1.99321

1.00000

2.00679
1.99321

1.00000

0.01359

1.99321
2.99304
1.99321
2.99304

7.00000
3.99995
2.99304
8.00692
0.02709
2.99304
5.04097

p = 0.00676, q = 0.00692.
p4 = p3 + p = 1.99996, q4 = q3 + q = 2.99996.

22.00000
15.95943
6.00642
0.03416
10.04768
0.04067
10.05419

24.00000
0.06808
23.96501
0.10307

249

## 250 Numerical Analysis

Fifth iteration
1.00000

4.00000
1.99996

1.00000

2.00004
1.99996

1.00000

0.00008

1.99996
2.99996
1.99996
2.99996

7.00000
4.00000
2.99996
8.00004
0.00015
2.99996
5.00023

22.00000
15.99978
6.00004
0.00018
10.00027
0.00023
10.00031

24.00000
0.00037
23.99981
0.00055

p = 0.00004, q = 0.00004.
p5 = p4 + p = 2.00000, q5 = q4 + q = 3.00000.
Therefore, a quadratic factor is x2 + 2x 3 which is equal to (x 1)(x + 3). The
deated polynomial is Q2 (x) = x2 + 2.00004x 8.00004  x2 + 2x 8.
Thus P4 (x) = (x 1)(x + 3)(x2 + 2x 8) = (x 1)(x + 3)(x 2)(x + 4).
Hence the roots of the given equation are 1, 3, 2, 4.
Algorithm 4.7 (Bairstow method). This algorithm extracts a quadratic factor from a polynomial of degree n and also determines the deated polynomial, by
Bairstow method.
Algorithm Bairstow
// Extract a quadratic factor x2 + px + q from a polynomial Pn (x) = xn + a1 xn1 +
+ an1 x + an of degree n and determines the deated polynomial Qn2 (x) =
xn2 + b1 xn3 + b2 xn4 + + bn2 .//
//the degree and the coecients of the polynomial.//
Read n, a1 , a2 , . . . , an ;
//the initial guess of p, q and error tolerance.//
Set b0 = 1, b1 = 0, c0 = 1, c1 = 0;
//Compute bk and ck
1. for k = 1 to n do
Compute bk = ak pbk1 qbk2 ;
endfor;
for k = 1 to n 1 do
Compute ck = bk pck1 qck2 ;
endfor;
(cn1 bn1 )bn cn2
cn3 bn1 cn2
;
q = bn1
Compute p = c2 bnc
c2n2 cn3 (cn1 bn1 ) ;
n3 (cn1 bn1 )
n2
Compute pnew = p + p, qnew = q + q;
if (|pnew p| < ) and (|qnew q| < ) then
Print The values of p and q are, pnew , qnew ;
Stop;
endif;
Set p = pnew , q = qnew ;

## Sol. of Algebraic and Transcendental Equs.

go to 1.
Print The coecients of deated polynomial are,b1 , b2 , . . . , bn2 ;
end Bairstow
Program 4.7
.
/* Program Bairstow for polynomial equation
Program to find all the roots of a polynomial equation by
Bairstow method. Leading coefficient is 1. Assume
initial guess for all p and q are 0.5, 0.5. */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
int n,i,k;
float p,q,pnew,qnew,a,b,c,bm1,cm1,delp,delq;
float epp=1e-5; /* error tolerance */
void findroots(float p, float q);
printf("\nEnter the degree of the polynomial ");
scanf("%d",&n);
printf("Enter the coefficients of the polynomial,
for(i=1;i<=n;i++) scanf("%f",&a[i]);
q=0.5;
p=0.5;
printf("The roots are \n");
do
{
b=1; bm1=0; c=1; cm1=0;
pnew=p; qnew=q;
do{
p=pnew; q=qnew;
b=a-p*b-q*bm1;
c=b-p*c-q*cm1;
for(k=2;k<=n;k++) b[k]=a[k]-p*b[k-1]-q*b[k-2];
for(k=2;k<=n;k++) c[k]=b[k]-p*c[k-1]-q*c[k-2];
delp=-(b[n]*c[n-3]-b[n-1]*c[n-2])/
(c[n-2]*c[n-2]-c[n-3]*(c[n-1]-b[n-1]));
delq=-(b[n-1]*(c[n-1]-b[n-1])-b[n]*c[n-2])/
(c[n-2]*c[n-2]-c[n-3]*(c[n-1]-b[n-1]));

251

## 252 Numerical Analysis

pnew=p+delp;
qnew=q+delq;
}while((fabs(pnew-p)>epp || fabs(qnew-q)>epp));
findroots(p,q);
n-=2;
for(i=1;i<=n;i++) a[i]=b[i];
}while(n>2);
/* deflated polynomial is quadratic */
if(n==2) findroots(b,b);
/* deflated polynomial is linear */
if(n==1) printf("%f ",-b);
} /* main */
/* finds the roots of the quadratic x*x+px+q=0 */
void findroots(float p, float q)
{
float dis;
dis=p*p-4*q;
if(dis>=0)
{
printf("%f %f\n",(-p+sqrt(dis))/2,(-p-sqrt(dis))/2);
}
else
{
printf("(%f,%f), (%f,%f)\n",-p/2,sqrt(fabs(dis))/2,
-p/2,-sqrt(fabs(dis))/2);
}
} /* findroots */
A sample of input/output:
Enter the degree of the polynomial 5
Enter the coefficients of the polynomial, except leading coeff.
3 4 -5 6 1
The roots are
(0.604068,0.729697), (0.604068,-0.729697)
(-2.030660,1.861934), (-2.030660,-1.861934)
-0.146816

253

Direct Method
4.16

## Graees Root Squaring Method

This method may be used to nd all the roots of all types (real, equal or complex) of a
polynomial equation with real coecients. In this method, an equation is constructed
whose roots are squares of the roots of the given equation, then another equation whose
roots are squares of the roots of this new equation and so on, the process of root-squaring
being continued as many times as necessary.
Let the given equation be
xn + a1 xn1 + a2 xn2 + + an1 x + an = 0

(4.88)

## whose roots are 1 , 2 , . . . , n .

Now, an equation is constructed whose roots are 12 , 22 , . . . , n2 using the following
technique.
Separating the even and odd powers of x in (4.88), and squaring both sides
(xn + a2 xn2 + a4 xn4 + )2 = (a1 xn1 + a3 xn3 + )2 .
After simplication the above equation becomes
x2n (a21 2a2 )x2n2 +(a22 2a1 a3 +2a4 )x2n4 + +(1)n a2n = 0.

(4.89)

## Setting x2 = z to the above equation and let it be

z n + b1 z n1 + + bn1 z + bn = 0
where

b1 = a21 2a2
b2 = a22 2a1 a3 + 2a4
..
.
bk = a2k 2ak1 ak+1 + 2ak2 ak+2
..
.

(4.90)

bn = a2n .
The roots of the equation (4.89) are 12 , 22 , . . . , n2 . The coecients bk s can be
obtained from Table 4.3.
The (k + 1)th column i.e., bk of Table 4.3 can be obtained as follows:
The terms alternate in sign starting with a positive sign. The rst term is a2k . The
second term is twice the product of ak1 and ak+1 . The third term is twice the product

## Table 4.3: Graees root-squaring scheme.

1
1

a1
a21
2a2

a2
a22
2a1 a3
2a4

a3
a23
2a2 a4
2a1 a5
2a6

b1

b2

b3

a4
a24
2a3 a5
2a2 a6
a1 a7
2a8
b4

an
a2n

bn

of ak2 and ak+2 . This process is continued until there are no available coecients to
form the cross product terms.
The root-squaring process is repeated to a sucient number of times, say m times
and we obtain the equation
xn + c1 xn1 + c2 xn2 + + cn1 x + cn = 0.

(4.91)

Let the roots of (4.91) be 1 , 2 , . . . , n . This roots are the 2m th power of the roots
of the equation (4.88) with opposite signs, i.e.,
m

i = i2 , i = 1, 2, . . . , n.
The relation between roots and coecients gives

 p
c1 = 1 + 2 + + n =
i =
i

 p p
c2 = 1 2 + 1 3 + + n1 n =
i j =
i j
c3 =


i<j<k

i j k =

i<j

i<j

ip jp kp

i<j<k

(1)n cn = 1 2 n = (1)n 1p 2p np , where p = 2m .
In the following dierent cases are considered separately.
Case I. Roots are real and unequal in magnitudes.
Let us consider
|n | < |n1 | < < |2 | < |1 |
then
|n |  |n1 |   |2 |  |1 |.

(4.92)

## Sol. of Algebraic and Transcendental Equs.

255

That is,
m

|n |2  |n1 |2   |2 |2  |1 |2

(4.93)

since all the roots are widely separated in magnitude at the nal stage.
Then from (4.92),

 p  p
 p

2
3
n
p
c1 = 1 1 +
 1p
+
+ + +
1
1
1
as

 p
i

1
Similarly,



 p

1 3
n1 n p
p
 (1 2 )p
+ +
c2 = (1 2 ) 1 +
1 2
1 2

## and so on, nally,

cn = (1 2 n )p .
Thus at the desired level of accuracy
1/p

## |1 | = c1 , |2 | = (c2 /c1 )1/p , |n | = (cn /cn1 )1/p , p = 2m .

(4.94)

This determines the absolute values of the roots. By substituting these values in the
original equation (4.88) one can determine the sign of the roots. The squaring process is
terminated when another squaring process produces new coecients that are almost the
squares of the corresponding coecients ck s i.e., when the cross product terms become
negligible with respect to square terms. Thus the nal stage is identied by the fact
that on root-squaring at that stage all the cross products will vanish.
Case II. All roots are real with one pair of equal magnitude.
Let 1 , 2 , . . . , n be the roots of the given equation, if a pair of roots are equal in
magnitude then this pair is conveniently called a double root. A double root can be
identied in the following way:
If the magnitude of the coecient ck is about half the square of the magnitude of the
corresponding coecient in the previous equation, then it indicates that k is a double
root. The double root is determined by the following process.
We have
ck
ck+1
k 
and
k+1 
.
ck1
ck
Then


c
 k+1 
k k+1  k2  
.
ck1

Therefore,
|k2 | = |k |2(2

m)


c
 k+1 
=
.
ck1

1/p

## |1 | = c1 , |2 | = (c2 /c1 )1/p , . . . , |k1 | = (ck1 /ck2 )1/p , . . . ,

|k | = (ck+1 /ck1 )1/(2p) , . . . , |n | = (cn /cn1 )1/p ,

(4.95)

where p = 2 .
This gives the magnitude of the double root. The sign is determined by substituting
the root to the equation.
The double root can also be determined directly since k and k+1 converge to the
same root after sucient squaring. Generally, the rate of convergence to the double
root is slow.
Case III. One pair of complex roots and other roots are distinct in magnitude.
Let k and k+1 form a complex pair and let
k , k+1 = k eik
where k = |k | = |k+1 |.
For suciently large m, k can be determined from the previous case,


c

 ck+1 
 k+1 
2p
2



|k |  
or,


, where p = 2m .
k
ck1 
ck1
and k is determined from the relation
2m
k cos mk 

ck+1
.
ck1

1/p

c

k+1

ck1

1/p
,

## |k+2 | = (ck+2 /ck+1 )1/p , . . . , |n | = (cn /cn1 )1/p , p = 2m .

The real roots 1 , 2 , . . . , k1 , k+2 , . . . , n are then corrected for sign.
If the equation has only one pair of complex roots k , k+1 = u iv then the sum of
the roots is
1 + 2 + + k1 + 2u + k+2 + + n = a1 .

## Sol. of Algebraic and Transcendental Equs.

257

From this relation one can determine the value of u. Then the value of v can be
determined from the relation
v 2 = 2k u2 .
The presence of complex roots in (k + 1)th column is identied by the following
technique:
If the coecients of xnk in the successive squaring to uctuate both in magnitude and
sign, a complex pair can be detected by this oscillation.
Merits and Demerits
1. All roots are found at the end of the method, i.e., at one execution of the method
all the roots are determined, including complex roots.
2. No initial guess is required.
3. As a direct method, there is no scope for correcting the error generated in any
stage. If any error is generated at any stage, then the error propagates to all the
subsequent computations and ultimately gives a wrong result.
4. The method is laborious, and to get a very accurate result the method has to be
repeated for a large number of times.
5. There is a chance for data overow in computer.
In the following, three examples are considered to discuss the three possible cases of
Graees method. The following table is the Graees root squaring scheme for four
degree equation.
1
1

a1
a2
a3
a4
a21
a22
a23
a24
2a2 2a1 a3 2a2 a4
2a4
1 c1
c2
c3
c4
Example 4.16.1 Find the roots of the equation
2x4 15x3 + 40x2 45x + 18 = 0
correct up to four decimal places by Graees root squaring method.
Solution. All the calculations are shown in the following table. The number within
the parenthesis represents the exponent of the adjacent number. i.e., 0.75(02) means
0.75 102 .

## 258 Numerical Analysis

m 2m
x4
x3
x2
0 1 1.00000 0.75000(01) 0.20000(02)
1.00000 0.56250(02) 0.40000(03)
0.40000(02) 0.33750(03)
0.18000(02)
1 2 1.00000 0.16250(02) 0.80500(02)
1.00000 0.26406(03) 0.64803(04)
0.16100(03) 0.47531(04)
0.16200(03)
2 4 1.00000 0.10306(03) 0.18891(04)
1.00000 0.10622(05) 0.35688(07)
0.37783(04) 0.17207(07)
0.13122(05)
3 8 1.00000 0.68436(04) 0.18612(07)
1.00000 0.46835(08) 0.34640(13)
0.37223(07) 0.61457(12)
0.86093(08)
4 16 1.00000 0.43113(08) 0.28495(13)
1.00000 0.18587(16) 0.81195(25)
0.56989(13) 0.16002(24)
0.37060(16)
5 32 1.0000 0.18530(16) 0.79595(25)

x
1
0.22500(02) 0.90000(01)
0.50625(03) 0.81000(02)
0.36000(03)
0.14625(03) 0.81000(02)
0.21389(05) 0.65610(04)
0.13041(05)
0.83481(04) 0.65610(04)
0.69690(08) 0.43047(08)
0.24789(08)
0.44901(08) 0.43047(08)
0.20161(16) 0.18530(16)
0.16023(15)
0.18559(16) 0.18530(16)
0.34443(31) 0.34337(31)
0.10560(29)
0.34337(31) 0.34337(31)

This is the nal equation since all the cross products vanish at the next step and all
the roots are real and distinct in magnitude.
Therefore,
|1 | = (0.18530 1016 )1/32 = 3.0000,


0.79595 1025 1/32
|2 | =
= 2.0000,
0.18530 1016

|3 | =

|4 | =

0.34337 1031
0.79595 1025
0.34337 1031
0.34337 1031

1/32
= 1.5000,
1/32
= 1.0000.

259

## Example 4.16.2 Find the roots of the equation x4 3x3 + 6x 4 = 0 correct up

to four decimal places by Graees root squaring method.
Solution. The necessary calculations are shown below.
m 2m
x4
x3
x2
0 1 1.00000 -0.30000(01) 0.00000(00)
1.00000 0.90000(01) 0.00000(00)
0.00000(00) 0.36000(02)
-0.80000(01)
1 2 1.00000 0.90000(01) 0.28000(02)
1.00000 0.81000(02) 0.78400(03)
-0.56000(02) -0.64800(03)
0.32000(02)
2 4 1.00000 0.25000(02) 0.16800(03)
1.00000 0.62500(03) 0.28224(05)
-0.33600(03) -0.20000(05)
0.51200(03)
3 8 1.00000 0.28900(03) 0.87360(04)
1.00000 0.83521(05) 0.76318(08)
-0.17472(05) -0.42763(08)
0.13107(06)
4 16 1.00000 0.66049(05) 0.33686(08)
1.00000 0.43625(10) 0.11347(16)
-0.67372(08) -0.57180(15)
0.85899(10)
5 32 1.00000 0.42951(10) 0.56296(15)

x
1
0.60000(01) -0.40000(01)
0.36000(02) 0.16000(02)
0.00000(00)
0.36000(02) 0.16000(02)
0.12960(04) 0.25600(03)
-0.89600(03)
0.40000(03) 0.25600(03)
0.16000(06) 0.65536(05)
-0.86016(05)
0.73984(05) 0.65536(05)
0.54736(10) 0.42950(10)
-0.11450(10)
0.43286(10) 0.42950(10)
0.18737(20) 0.18447(20)
-0.28936(18)
0.18447(20) 0.18447(20)

The diminishing double products vanish at the next step and hence this is the nal
equation and since we nd the characteristic behaviour of a double root in the second
column.
Therefore,
|1 | = (0.42951 1010 )1/32 = 2.0000,


0.18447 1020 1/64
|2 | = |3 | =
= 1.4142,
0.42951 1010


0.18447 1020 1/32
|4 | =
= 1.0000.
0.18447 1020
Here 1.4142 as well as -1.4142 satised the given equation, hence the roots of the
given equation are 2, 1.4142, 1.

## Example 4.16.3 Find the roots of the equation x4 5x3 + x2 7x + 10 = 0 correct

up to four decimal places by Graees root squaring method.
Solution. The necessary calculations are shown below.
m 2m
x4
x3
x2
0 1 1.00000 -0.50000(01) 0.10000(01)
1.00000 0.25000(02) 0.10000(01)
-0.20000(01) -0.70000(02)
0.20000(02)
1 2 1.00000 0.23000(02) -0.49000(02)
1.00000 0.52900(03) 0.24010(04)
0.98000(02) -0.13340(04)
0.20000(03)
2 4 1.00000 0.62700(03) 0.12670(04)
1.00000 0.39313(06) 0.16053(07)
-0.25340(04) -0.13344(08)
0.20000(05)
3 8 1.00000 0.39060(06) -0.11719(08)
1.00000 0.15256(12) 0.13732(15)
0.23437(08) -0.68659(14)
0.20000(09)
4 16 1.00000 0.15259(12) 0.68665(14)
1.00000 0.23283(23) 0.47148(28)
-0.13733(15) -0.30727(28)
0.20000(17)
5 32 1.00000 0.23283(23) 0.16422(28)

x
1
-0.70000(01) 0.10000(02)
0.49000(02) 0.10000(03)
-0.20000(02)
0.29000(02) 0.10000(03)
0.84100(03) 0.10000(05)
0.98000(04)
0.10641(05) 0.10000(05)
0.11323(09) 0.10000(09)
-0.25340(08)
0.87891(08) 0.10000(09)
0.77248(16) 0.10000(17)
0.23437(16)
0.10069(17) 0.10000(17)
0.10137(33) 0.10000(33)
-0.13733(31)
0.10000(33) 0.10000(33)

## Since c2 alternates in sign, it indicates that there is a pair of complex roots.

|1 | = (.23283 1023 )1/32 = 5.0000,


0.10000 1033 1/64
|4 | =
= 1.0000.
0.10000 1033
These two roots are positive.

22

0.10000 1033
0.23283 1023

1/32
= 2.0000.

If 2 
, 3 = u v, then 2u + 5 + 1 = 5 (sum of the roots). Therefore, u = 0.5. Then

v = 22 u2 = 2 0.25 = 1.3229.
Hence the roots are 5, 1, 0.5 1.3229i.

4.17

261

## Solution of Systems of Nonlinear Equations

To solve a system of nonlinear equations the following methods are discussed in this
section.
1. The method of iteration (xed point iteration)
2. Seidal iteration
3. Newton-Raphson method.
4.17.1

## Let the system of nonlinear equations be

f (x, y) = 0
and

g(x, y) = 0.

(4.96)

whose real roots are required within a specied accuracy. The above system can be
rewritten as
x = F (x, y)
and

y = G(x, y).

(4.97)

## The function F and G may be obtained in many dierent ways.

Let (x0 , y0 ) be the initial guess to a root (, ) of the system (4.96). Then we obtain
the following sequence {(xn , yn )} of roots.
x1 = F (x0 , y0 ),

y1 = G(x0 , y0 )

x2 = F (x1 , y1 ),

y2 = G(x1 , y1 )

(4.98)

## xn+1 = F (xn , yn ), yn+1 = G(xn , yn ).

If the sequence (4.98) converges, i.e.,
lim xn =

and

lim yn =

then
= F (, )

and

= G(, ).

(4.99)

Like the iteration process for single variable, the above sequence surely converge to
a root under certain condition. The sucient condition is stated below.

## 262 Numerical Analysis

Theorem 4.12 Assume that the functions x = F (x, y), y = G(x, y) and their rst
order partial derivatives are continuous on a region R that contains a root (, ). If the
starting point (x0 , y0 ) is suciently close to (, ) and if
 

 



 F   F 
 G   G 
+
<1





and
(4.100)
 x   y 
 x  +  y  < 1,
for all (x, y) R, then the iteration scheme (4.98) converges to a root (, ).
The condition for the functions x = F (x, y, z), y = G(x, y, z), z = H(x, y, z) is
 
 


 F   F   F 
 
 


+
+
 < 1,

 x   y   z 
 
 


 G   G   G 
 
 


+
+
<1

 x   y   z 
 
 


 H   H   H 
 
 


and 
+
+
<1
 x   y   z 
for all (x, y, z) R.
4.17.2

Seidal method

An improvement of the iteration method can be made by using the recently computed
values of xi while computing yi , i.e., xi+1 is used in the calculation of yi . Therefore, the
iteration scheme becomes
xn+1 = F (xn , yn )
yn+1 = G(xn+1 , yn ).

(4.101)

This method is called Seidal iteration. In case of three variables the scheme is
xn+1 = F (xn , yn , zn )
yn+1 = G(xn+1 , yn , zn )

(4.102)

## and zn+1 = H(xn+1 , yn+1 , zn ).

Example 4.17.1 Solve the following system of equations
x=

2x x2 + 4y y 2 + 3
8x 4x2 + y 2 + 1
and y =
8
4

starting with (x0 , y0 ) = (1.1, 2.0), using (i) iteration method, and (ii) Seidal iteration
method.

## Sol. of Algebraic and Transcendental Equs.

263

Solution.
(i) Iteration method
Let
F (x, y) =

8x 4x2 + y 2 + 1
8

and

G(x, y) =

2x x2 + 4y y 2 + 3
.
4

## The iteration scheme is

8xn 4x2n + yn2 + 1
,
8
2xn x2n + 4yn yn2 + 3
.
= G(xn , yn ) =
4

xn+1 = F (xn , yn ) =
and yn+1

The value of xn , yn , xn+1 and yn+1 for n = 0, 1, . . . are shown in the following table.
n
0
1
2
3
4

xn
1.10000
1.12000
1.11655
1.11641
1.11653

yn
2.00000
1.99750
1.99640
1.99660
1.99661

xn+1
1.12000
1.11655
1.11641
1.11653
1.11652

yn+1
1.99750
1.99640
1.99660
1.99661
1.99660

## Therefore, a root correct up to four decimal place is (1.1165, 1.9966).

(ii) Seidal method
The iteration scheme for Seidal method is
8xn 4x2n + yn2 + 1
8
2xn+1 x2n+1 + 4yn yn2 + 3
.
= G(xn+1 , yn ) =
4

xn+1 = F (xn , yn ) =
and

yn+1

n
0
1
2
3

xn
1.10000
1.12000
1.11600
1.11659

yn
2.00000
1.99640
1.99663
1.99660

xn+1
1.12000
1.11600
1.11659
1.11650

## Therefore, root correct up to four decimal places is

(1.1165, 1.9966).

yn+1
1.99640
1.99663
1.99660
1.99660

## 264 Numerical Analysis

Algorithm 4.8 (Seidal iteration). This algorithm used to solve two non-linear
equations by Seidal iteration, when the initial guess is given.
Algorithm Seidal-Iteration-2D
// Let (x0 , y0 ) be the initial guess of the system of equations x = F (x, y), y = G(x, y).
be the error tolerance, maxiteration represents the maximum number of repetitions
to be done.//
Input functions F (x, y), G(x, y).
Read , maxiteration, x0 , y0 , z0 ;
Set k = 0, error = 1;
While k < maxiteration and error > do
Set k = k + 1;
Compute x1 = F (x0 , y0 ), y1 = G(x1 , y0 );
Compute error = |x1 x0 | + |y1 y0 |;
Set x0 = x1 , y0 = y1 ;
endwhile;
if error < then
Print The sequence converge to the root, x1 , y1 ;
Stop;
else
Print The iteration did not converge after, k,iterations;
Stop;
endif;
end Seidal-Iteration-2D
Program 4.8
.
/* Program Seidal for a pair of non-linear equations
Program to find a root of a pair of non-linear equations
by Seidal method. Assumed that the equations are given
in the form x=f(x,y) and y=g(x,y).
The equations taken are x*x+4y*y-4=0, x*x-2x-y+1=0. */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
int k=0,maxiteration;
float error=1,eps=1e-5,x0,y0; /*initial guesses for x and y*/
float x1,y1;
float f(float x, float y);

## float g(float x, float y);

printf("Enter initial guess for x and y ");
scanf("%f %f",&x0,&y0);
printf("Maximum iterations to be allowed ");
scanf("%d",&maxiteration);
while((k<maxiteration) && (error>eps))
{
k++;
x1=f(x0,y0); y1=g(x1,y0);
error=fabs(x1-x0)+fabs(y1-y0);
x0=x1; y0=y1;
}
if(error<eps)
{
printf("The sequence converges to the\n");
printf("root (%7.5f, %7.5f) at %d iterations",x1,y1,k);
exit(0);
}
else
{
printf("The iteration did not converge after %d iterations",k);
exit(0);
}
} /* main */
/* definition of f(x,y) */
float f(float x, float y)
{
float f1;
f1=sqrt(4-4*y*y);
return f1;
}
/* definition of g(x,y) */
float g(float x, float y)
{
float g1;
g1=x*x-2*x+1;
return g1;
}

265

## 266 Numerical Analysis

A sample of input/output:
Enter initial guess for x and y 0.5 0.5
Maximum iterations to be allowed 100
The sequence converges to the root (0.00000, 1.00000) at 68 iterations
4.17.3

Newton-Raphson method

## Let (x0 , y0 ) be an initial guess to the root (, ) of the equations

f (x, y) = 0 and g(x, y) = 0.

(4.103)

## If (x0 + h, y0 + k) is the root of the above system then

f (x0 + h, y0 + k) = 0
g(x0 + h, y0 + k) = 0.

(4.104)

## Assume that f and g are dierentiable. Expanding (4.104) by Taylors series.

'
& '
f
f
f (x0 , y0 ) + h
+k
+ = 0
x
y
(x ,y )
(x ,y )
& ' 0 0
& ' 0 0
g
g
+k
+ = 0
g(x0 , y0 ) + h
x
y
&

(x0 ,y0 )

(4.105)

(x0 ,y0 )

Neglecting square and higher order terms, the above equations simplied as
f0
f0
+k
= f0
x
y
g0
g0
+k
= g0
h
x
y

f0  f 
=
etc.
x
x (x0 ,y0 )
The above system can be written as

where f0 = f (x0 , y0 ),

f
0
x

g
0
x

f0

y h
f

0
1 f0
=
or
=
J
.

g0
g0
k
g k
0

(4.106)

## Alternatively, h and k can be evaluated as




f0 

 f0
 f0


f0


y
1
1  x

h= 
, k = 

J
J  g
 g g0 
 0 g



0
0
y
x




 f0



 x


 , where J = 

 g

 0


x


f0 

y 

.
g0 

y

267

(4.107)

Thus h and k are determined by one of the above two ways. Therefore, the new
approximations are then given by
x1 = x0 + h,

y1 = y0 + k.

(4.108)

The process is to be repeated until the roots are achieved to the desired accuracy.
The general formula is xn+1 = xn + h, yn+1 = yn + k; h, k are evaluated at (xn , yn )
instead at (x0 , y0 ).
If the iteration converges (the condition is stated below) then the rate of convergence
Theorem 4.13 Let (x0 , y0 ) be an initial guess to a root (, ) of the system f (x, y) =
0, g(x, y) = 0 in a closed neighbourhood R containing (, ). If
1. f, g and their rst order partial derivatives are continuous and bounded in R, and
2. J = 0 in R, then the sequence of approximation xn+1 = xn + h, yn+1 = yn + k,
where h and k are given by (4.107), converges to the root (, ).
Example 4.17.2 Use Newton-Raphson method to solve the system x2 2x y +
0.5 = 0, x2 + 4y 2 4 = 0 with the starting value (x0 , y0 ) = (2.00, 0.25).
Solution. Let f (x, y) = x2 2x y + 0.5 and g(x, y) = x2 + 4y 2 4.
f
g
g
f
= 2x 2,
= 1,
= 2x,
= 8y.
x
y
x
y
Therefore,
(
J=

At (x0 , y0 ), J0 =
Therefore,

f
x
g
x

f
y
g
y

2x 2 1
,
2x
8y

f0
g0

2 1
.
4 2

2 1
4 2

h
k

0.25
0.25

0.25
.
0.25

or,

h
k

1
=
8

2 1
4 2

0.25
0.25

0.09375
0.06250

## Thus, x1 = x0 + h = 2.00 0.09375 = 1.90625,

y1 = y0 + k = 0.25
+ 0.0625 = 0.31250.

1.81250 1.00000
f1
0.00879
,
At (x1 , y1 ), J1 =
=
.
g1
3.81250 2.50000
0.02441

h
f1
=
J1
k
g1

1
2.50000 1.00000
h
0.00879
0.00556
or,
=
=
.
k
0.02441
0.00129
8.34375 3.81250 1.81250
Therefore, x2 = x1 + h = 1.90625 0.00556 = 1.90069,
0.00129 = 0.31121.
y2 = y1 + k = 0.31250

0.00003
1.80138 1.00000
f2
=
.
,
At (x2 , y2 ), J2 =
0.00003
g2
3.80138 2.48968

h
f2
=
J2
k
g2

1
h
2.48968 1.00000
0.00003
0.00001
or,
=
=
.
k
0.00003
0.00001
8.28624 3.80138 1.80138
Hence, x3 = x2 + h = 1.90069 0.00001 = 1.90068,
y3 = y2 + k = 0.31121 + 0.00001 = 0.31122.
Thus, one root is x = 1.9007, y = 0.3112 correct up to four decimal places.
Algorithm 4.9 (Newton-Raphson method for pair of equations). This algorithm solves a pair of non-linear equations by Newton-Raphson method. The initial
guess of a root is to be supplied.
Algorithm Newton-Raphson -2D
//(x0 , y0 ) is initial guess, is the error tolerance.//
Input functions f (x, y), g(x, y), fx (x, y), fy (x, y), gx (x, y), gy (x, y).
Read x0 , y0 , , maxiteration;
for i = 1 to maxiteration do
Compute f0 = f (x0 , y0 ), g0 = g(x0 , y0 );
if (|f0 | < and |g0 | < ) then
Print A root is, x0 , y0 ;
Stop;
endif;
 
 
Compute delf x =

f
x

,
(x0 ,y0 )

delf y =

f
y

(x0 ,y0 )




Compute delgx =

g
x


, delgy =

(x0 ,y0 )

g
y

(x0 ,y0 )

## Compute J = delf x delgy delgx delf y;

Compute h = (f0 delgy + g0 delf y)/J;
Compute k = (g0 delf x + f0 delgx)/J;
Compute x0 x0 + h, y0 y0 + k;
endfor;
Print Solution does not converge in, maxiteration, iteration;
end Newton-Raphson -2D
Program 4.9
.
/* Program Newton-Raphson (for a pair of non-linear equations)
Program to find a root of a pair of non-linear equations
by Newton-Raphson method. Partial derivative of f and g
w.r.t. x and y are to be supplied.
The equations taken are 3x*x-2y*y-1=0, x*x-2x+2y-8=0. */
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
int i,maxiteration;
float eps=1e-5,x0,y0; /* initial guesses for x and y */
float J,k,h,delfx,delfy,delgx,delgy,f0,g0;
float f(float x, float y);
float fx(float x, float y);
float fy(float x, float y);
float g(float x, float y);
float gx(float x, float y);
float gy(float x, float y);
printf("Enter initial guess for x and y ");
scanf("%f %f",&x0,&y0);
printf("Maximum iterations to be allowed ");
scanf("%d",&maxiteration);
for(i=1;i<=maxiteration;i++)
{
f0=f(x0,y0);
g0=g(x0,y0);
if(fabs(f0)<eps && fabs(g0)<eps){

269

## printf("One root is (%7.5f, %7.5f) obtained at %d iterations",

x0,y0,i);
exit(0);
}
delfx=fx(x0,y0); delfy=fy(x0,y0);
delgx=gx(x0,y0); delgy=gy(x0,y0);
J=delfx*delgy-delgx*delfy;
h=(-f0*delgy+g0*delfy)/J;
k=(-g0*delfx+f0*delgx)/J;
x0+=h; y0+=k;
}
printf("Iteration does not converge in %d iterations",
maxiteration);
} /* main */
/* definition of f(x,y) and its partial derivative w.r.t x and y
fx(x,y) and fy(x,y) */
float f(float x, float y)
{
return (3*x*x-2*y*y-1);
}
float fx(float x, float y)
{
return (6*x);
}
float fy(float x, float y)
{
return (-4*y);
}
/*definition of g(x,y) and its partial derivative w.r.t x and y
gx(x,y) and gy(x,y) */
float g(float x, float y)
{
return (x*x-2*x+2*y-8);
}
float gx(float x, float y)
{
return (2*x-2);
}

271

## float gy(float x, float y)

{
return (2);
}
A sample of input/output:
Enter initial guess for x and y 2.5 3
Maximum iterations to be allowed 50
One root is (2.64005, 3.15512) obtained at 4 iterations

4.18

Exercise

## 1. Find the number of real roots of the equation (i) 2x 3 = 0,

(ii) x10 4x4 100x + 200 = 0.
2. Describe the methods to locate the roots of the equation f (x) = 0.
3. Obtain a root for each of the following equations using bisection method, regulafalsi method, iteration method, Newton-Raphson method, secant method.
(i) x3 + 2x2 x + 7 = 0, (ii) x3 4x 9 = 0, (iii) cos x = 3x 1
(iv) ex = 10x, (v) sin x = 10(x 1), (vi) sin2 x = x2 1
(vii) tan x tanh x = 0, (viii) x3 + 0.5x2 7.5x 9.0 = 0, (ix) tan x + x = 0,
(x) x3 5.2x2 17.4x + 21.6 = 0, (xi) x7 + 28x4 480 = 0,
(xii) (x 1)(x 2)(x 3) = 0, (xiii) x cos x = 0, (xiv) x + log x = 2,
(xv) sin x = 12 x, (xvi) x3 sin x + 1 = 0, (xvii) 2x = cos x + 3,
sin x = 0, (xx) 2x 2x2 1 = 0,
(xviii) x log10 x = 4.77, (xix) x2
x
(xxi) 2 log x 2 + 1 = 0, (xxii) 1 + x = 1/x, (xxiii) log x + (x + 1)3 = 0.
4. Find a root of the equation x = 12 + sin x by using the xed point iteration method
xn+1 =

1
+ sin xn , x0 = 1
2

## correct to six decimal places.

5. Use Newton-Raphson method for multiple roots to nd the roots of the following
equations.
(i) x3 3x + 2 = 0, (ii) x4 + 2x3 2x 1 = 0.
6. Describe the bisection method to nd a root of the equation f (x) = 0 when
f (a) f (b) < 0, a, b be two specied numbers. Is this condition necessary to get a
root using this method ?

## 272 Numerical Analysis

7. Describe regula-falsi method for nding a real root of an equation. Why this
method is called the bracketing method ? Give a geometric interpretation of this
method. What is the rate of convergence of regula-falsi method ? Compare this
of this method.
8. Compare bisection method and regula-falsi method. Also compare bracketing
methods and iterative methods.
9. What is the main dierence between regula-falsi method and secant method ?
10. Explain how an equation f (x) = 0 can be solved by the method of iteration (xed
point iteration) and deduce the condition of convergence. Show that the rate of
convergence of iteration method is one. Give a geometric interpretation of this
11. Find the iteration schemes to solve the following equations using xed point iteration method:
(i) 2x sin x 1 = 0, (ii) x3 2x 1 = 0, (iii) x + sin x = 0.
12. Describe Newton-Raphson method for computing a simple real root of an equation f (x) = 0. Give a geometrical interpretation of the method. What are the
method has a second order convergence.
13. Find the iteration schemes to solve the following equations using Newton-Raphson
method:
(i) 2x cos x 1 = 0, (ii) x5 + 3x2 1 = 0, (iii) x2 2 = 0.
14. Use Newton-Raphson
method
to nd the value of the following terms:

3
(i) 1/15, (ii) 11, (iii) 5.
15. Show that an iterative method for computing
xn+1
and also deduce that

a is given by

1
a
(k 1)xn + k1
=
k
xn
k1 2
n+1 

2ka n

where n is the error at the nth iteration. What is the order of convergence for
this iterative method ?

273

## 16. Use Newton-Raphson method and modied Newton-Raphson method to nd the

roots of the following equations and compare the methods.
(i) x3 3x + 2 = 0, (ii) x4 + 2x3 2x 1 = 0.
17. Solve the following equation using Newton-Raphson method:
(i) z 3 4iz 2 3ez = 0, z0 = 0.53 0.36i, (ii) 1 + z 2 + z 3 = 0, z0 = 1 + i.
18. Compare Newton-Raphson method and modied Newton-Raphson method to nd
a root of the equation f (x) = 0.
19. Use modied Newton-Raphson method to nd a root of the equation
% x
1
2
et dt =
10
0
with six decimal places.
20. Device a scheme to nd the value of
21. The formula
xn+1 = xn

2
f  (xn )

## is used to nd a multiple root of multiplicity two of the equation f (x) = 0. Show

that the rate of convergence of this method is cubic.
22. The iteration scheme
xn+1 = xn

3 loge xn exn
p

## is used to nd the root of the equation ex 3 loge x = 0. Show that p = 3 gives

rapid convergence.
23. Use Mullers method to nd a root of the equations
(i) x3 2x 1 = 0, (ii) x4 6x2 + 3x 1 = 0.
24. Determine the order of convergence of the iterative method
xn+1 =

x0 f (xn ) xn f (x0 )
f (xn ) f (x0 )

## for nding the simple root of the equation f (x) = 0.

25. Find the order of convergence of the Steensen method
xn+1 = xn

f (xn )
f (xn + f (xn )) f (xn )
, g(xn ) =
, n = 0, 1, 2, . . . .
g(xn )
f (xn )

## 274 Numerical Analysis

26. Show that the iteration scheme to nd the value of a using Chebyshev third
order method is given by




1
a
a 2
1
.
xn+1 =

xn
xn +
2
xn
8xn
xn

## Use this scheme to nd the value of 17.

27. An iteration scheme is given by
x0 = 5,

xn+1 =

1 4 1 3
x x + 8xn 12.
16 n 2 n

## Show that it gives cubic convergence to = 4.

28. Using Graees root squaring method, nd the roots of the following equations
(i) x3 4x2 + 3x + 1 = 0, (ii) x3 3x 5 = 0,
(iii) x4 2x3 + 1.99x2 2x + 0.99 = 0, (iv) x3 + 5x2 44x 60 = 0.
29. Use Birge-Vieta method to nd the roots of the equations to 3 decimal places.
(i) x4 8x3 + 14.91x2 + 9.54x 25.92 = 0, (ii) x3 4x + 1 = 0,
(iii) x4 1.1x3 0.2x2 1.2x + 0.9 = 0.
30. Find the quadratic factors of the following polynomial equations using Bairstows
method.
(i) x4 8x3 + 39x2 62x + 50 = 0, (ii) x3 2x2 + x 2 = 0,
(iii) x4 6x3 + 18x2 24x + 16 = 0, (iv) x3 2x + 1 = 0.
31. Solve the following systems of nonlinear equations using iteration method
(i) x2 + y = 11, y 2 + x = 7, (ii) 2xy 3 = 0, x2 y 2 = 0.
32. Solve the following systems of nonlinear equations using Seidal method
(i) x2 + 4y 2 4 = 0, x2 2x y + 1 = 0, start with (1.5, 0.5),
(ii) 3x2 2y 2 1 = 0, x2 2x + y 2 + 2y 8 = 0, start with (1, 1).
33. Solve the following systems of nonlinear equations using Newton-Raphson method
(i) 3x2 2y 2 1 = 0, x2 2x + 2y 8 = 0 start with initial guess (2.5,3),
(ii) x2 x + y 2 + z 2 5 = 0, x2 + y 2 y + z 2 4 = 0, x2 + y 2 + z 2 + z 6 = 0
34. Use Newtons method to nd all nine solutions to 7x3 10x y 1 = 0, 8y 3
11y + x 1 = 0. Use the starting points (0, 0), (1, 0), (0, 1), (1, 0), (0, 1), (1, 1),
(1, 1), (1, 1) and (1, 1).
35. What are the dierences between direct method and iterative method ?

Chapter 5

## Solution of System of Linear

Equations
A system of m linear equations in n unknowns (variables) is written as
a11 x1 + a12 x2 + + a1n xn = b1

(5.1)

## The quantities x1 , x2 , . . ., xn are the unknowns (variables) of the system and

a11 , a12 , . . ., amn are the coecients of the unknowns of the system. The numbers
b1 , b2 , . . . , bm are constant or free terms of the system.
The above system of equations (5.1) can be written as
n

aij xj = bi ,
i = 1, 2, . . . , m.
(5.2)
j=1

where

A=

a11
a21

ai1

am1

a12
a22

ai2

am2

AX = b,

b1
x1
a1n
b2
x2

a2n

..
..

. and X = . .
,
b
=
bi
xi
ain

..

.
.
.
amn
bm
xm
275

(5.3)

(5.4)

## 276 Numerical Analysis

The system of linear equation (5.1) is consistent if it has a solution. If a system
of linear equations has no solution, then it is inconsistent (or incompatible). A
consistent system of linear equations may have one solution or several solutions and is
said to be determinate if there is one solution and indeterminate if there are more
than one solution.
Generally, the following three types of the elementary transformations to a system
of linear equations are used.
Interchange: The order of two equations can be changed.
Scaling: Multiplication of both sides of an equation of the system by any non-zero
number.
Replacement: Addition to (subtraction from) both sides of one equation of the corresponding sides of another equation multiplied by any number.
A system in which the constant terms b1 , b2 , . . . , bm are zero is called a homogeneous
system.
Two basic techniques are used to solve a system of linear equations:
(i) direct method, and (ii) iteration method.
Several direct methods are used to solve a system of equations, among them following
are most useful.
(i) Cramers rule, (ii) matrix inversion, (iii) Gauss elimination, (iv) decomposition, etc.
The most widely used iteration methods are (i) Jacobis iteration, (ii) Gauss-Seidals
iteration, etc.

Direct Methods
5.1

Cramers Rule

To solve a system of linear equations, a simple method (but, not ecient) was discovered
by Gabriel Cramer in 1750.
Let the determinant of the coecients of the system (5.2) be D = |aij |; i, j =
1, 2, . . . , n, i.e., D = |A|. In this method, it is assumed that D = 0. The Cramers
rule is described in the following. From the properties of determinant

 

 a11 a12 a1n   x1 a11 a12 a1n 

 

 a21 a22 a2n   x1 a21 a22 a2n 




x1 D = x1 
 =  


 

 an1 an2 ann   x1 an1 an2 ann 


 a11 x1 + a12 x2 + + a1n xn a12 a1n 


 a21 x1 + a22 x2 + + a2n xn a22 a2n  [Using the operation


= 

 C1 = C1 + x2 C2 + + xn Cn .]

 an1 x1 + an2 x2 + + ann xn an2 ann 


 b1 a12

 b a
=  2 22

 bn an2
= Dx1 (say).

277


a1n 
a2n 
[Using (5.1)]

ann 

## Therefore, x1 = Dx1 /D.

Dx2
Dxn
, . . . , xn =
.
Similarly, x2 =
D
D
Dxi
, where
In general, xi =
D

 a11 a12 a1 i1

 a a a2 i1
Dxi =  21 22

 an1 an2 an i1

b1 a1 i+1
b2 a2 i+1

bn an i+1


a1n 
a2n 
,

ann 

i = 1, 2, , n
Example 5.1.1 Use Cramers rule to solve the following systems of equations
x1 + x2 + x3 = 2
2x1 + x2 x3 = 5
x1 + 3x2 + 2x3 = 5.
Solution. The determinant D of the system is


1 1 1


D =  2 1 1  = 5.
1 3 2
The determinants D1 , D2 and D3 are shown below:




2 1 1
1 2 1




D1 =  5 1 1  = 5,
D2 =  2 5 1  = 10,
5 3 2
1 5 2



1 1 2


D3 =  2 1 5  = 5.
1 3 5

5
10
5
D1
D2
D3
= = 1, x2 =
=
= 2, x3 =
= = 1.
D
5
D
5
D
5
Therefore the solution is x1 = 1, x2 = 2, x3 = 1.
Thus, x1 =

5.1.1

## Computational aspect of Cramers rule

It may be noted that the Cramers rule involves to compute (n + 1) determinants each
of order n (for a system of n equations and n variables). Again, the Laplaces expansion

## 278 Numerical Analysis

method (the conventional method to nd the value of a determinant) to evaluate a
determinant of order n requires (n! 1) additions and n!(n 1) multiplications. Thus,
to compute a determinant of order 10, needs 32 million multiplications and 4 millions
additions. So, the Cramers rule is only of theoretical interest due to the computational
ineciency. But, the time complexity can be reduced by triangularizing the determinant
and this method is a polynomial time bound algorithm.

5.2

Evaluation of Determinant

## Triangularization, is also known as Gauss reduction method, is the best way to

evaluate a determinant. The basic principle of this method is to convert the given
determinant into a lower or upper triangular form by using only elementary row operations. If the determinant is reduced to a triangular form (D ) then the value of D is
obtained by multiplying the diagonal
elements of D .

 a11 a12 a1n 


 a21 a22 a2n 
 be a determinant of order n.

Let D = 

 
 an1 an2 ann 
Using the elementary row operations, D can be reduced to the following form:

 a11

 0


D =  0



 0

a12
(1)
a22
0

a13
(1)
a23
(2)
a33


a1n 
(1)
a2n 
(2) 
a3n  ,



(n1)
ann 

where
(k1)

(k)

(k1)

aij = aij

aik

(k1)
a
;
(k1) kj
akk

(5.5)

(0)

(1) (2)

(n1)

## Then the value of D is equal to a11 a22 a33 ann

To compute the value

(k)
(k1)
is zero
of aij one division is required. If akk
(k1)
If akk is small then the division leads to the loss

then further

of signicant
reduction is not possible.
digits. To prevent the loss of signicant digits, the pivoting techniques are used.
A pivot is the largest magnitude element in a row or a column or the principal
diagonal or the leading or trailing sub-matrix of order i (2 i n).

## For example, for the matrix

5
10
A=
20
3

279

1 0 5
8 3 10

3 30 8
50 9 10

20 is the pivot for the rst column, 30 is the pivot for the principal diagonal,
50
is the

5 1
pivot for this matrix and 10 is the pivot for the trailing sub-matrix
.
10 8
In the elementary row operation, if the any one of the pivot element is zero or very
small relative to other elements in that row then we rearrange the remaining rows in
such a way that the pivot becomes non-zero or not a very small number. The method is
called pivoting. The pivoting are of two types- partial pivoting and complete pivoting.
Partial and complete pivoting are discussed in the following.
Partial pivoting
In the rst stage, the rst pivot is determined by nding the largest element in magnitude among the elements of rst column and let it be ai1 . Then rows i and 1 are
interchanged. In the second stage, the second pivot is determined by nding the largest
element in magnitude among the elements of second column leaving rst element and
let it be aj2 . The second and jth rows are interchanged. This process is repeated for
(n 1)th times. In general, at ith stage, the smallest index j is chosen for which
(k)

(k)

(k)

(k)

(k)

## |aij | = max{|akk |, |ak+1 k |, . . . , |ank |} = max{|aik |, i = k, k + 1, . . . , n}

and the rows k and j are interchanged.
Complete pivoting or full pivoting
The largest element in magnitude is determined among all the elements of the determinant and let it be |alm |.
Taking alm as the rst pivot by interchanging rst row and the lth row and of rst
column and mth column. In second stage, the largest element in magnitude is determined among all elements leaving the rst row and rst column and taking this element
as second pivot.
In general, we choose l and m for which
(k)

(k)

## |alm | = max{|aij |, i, j = k, k + 1, . . . , n}.

Then the rows k, l and columns k, m are interchanged, and akk becomes a pivot.
The complete pivoting is more complicated than the partial pivoting. The partial
pivoting is preferred for hand computation.

## 280 Numerical Analysis

Note 5.2.1 If the coecient matrix A is diagonally dominant, i.e.,
n

j=1
j=i

or

n


## |aji | < |aii |,

for i = 1, 2, . . . , n.

(5.6)

j=1
j=i

## or real symmetric and positive denite then no pivoting is necessary.

Note 5.2.2 Every diagonally dominant matrix is non-singular.
To illustrate the partial pivoting, let us consider the matrix

1 7 3
A = 4 5 1.
8 1 6
The largest element (in magnitude) in the rst column is 8. Then interchanging
rst and third rows i.e.,

8 1 6
A 4 5 1.
1 7 3
The largest element in second column leaving the rst row is 7, so interchanging
second and third rows.
The matrix after partial pivoting is

8 1 6
A 1 7 3.
4 5 1

1 7 3
Let us consider the matrix B = 4 8 5 to illustrate the complete pivoting. The
2 6 1
largest element (in magnitude) is determined among all the elements of the matrix. It
is 8 attained at the (2, 2) position. Therefore, rst and second columns, and rst and
second rows are interchanged. The matrix transferred to

8 4 5
B 7 1 3.
6 2 1
The number 8 is the rst pivot.
To nd second pivot, largest element is determined
1 3
from the trailing sub-matrix
. The largest element is 3 and it is at the position
2 1
(2, 3).

## Solution of System of Linear Equations

281

Interchanging
second and third columns. The nal matrix (after complete pivoting)

8 5 4
is 7 3 1 .
6 1 2

Example 5.2.1 Compute the determinant of the following matrix by a triangularization algorithm using (i) partial pivoting, and (ii) complete pivoting:

2 0 4
A = 4 6 1.
5 1 2
Solution. (i) The largest element in the rst column is 5, which is the rst pivot of
A.
Interchanging rst and third rows, we obtain

5 1 2
4 6 1.
2 0 4
and sign = 1.
2
4
Adding times the rst row to the second row, times the rst row to the third
5
5
4
2


row i.e., R2 = R2 R1 and R3 = R3 R1 , we get
5
5

5 1
2
0 26/5 13/5 .
0 2/5 24/5
26
, which is the largest element (in magnitude) among the
5
elements of second column except rst row. Since this element is in the (2,2) position,
so no interchange of rows is required.
2/5
2/5
times the second row to the third row i.e., R3 = R3 +
R2 we obtain
26/5
26/5

5 1
2
0 26/5 13/5 .
0 0
5
The second pivot is

## 282 Numerical Analysis

(ii) The largest element in A is 6. Interchanging rst and second columns and setting
sign = 1; and then interchanging rst and second rows and setting sign = sign =
1, we have

6 4 1
0 2 4.
1 5 2
Adding 16 times the rst row to the third row i.e., R3 = R3 16 R1 we obtain

6 4
1
0 2
4 .
0 13/3 13/6

2
4
The pivot of the trailing sub-matrix
is 13/3. Interchanging the second
13/3 13/6

6 1
4
and third rows, we have 0 13/3 13/6 and sign = 1.
0 2
4
6
2
times the second row to the third row i.e., R3 = R3 R2 we obtain
13/3
13

6 1
4
0 13/3 13/6 .
0 0
5
Therefore, |A| = sign (6)(13/3)(5) = 130.
The algorithm for triangularization i.e., to nd the value of a determinant using
partial and complete pivoting are presented below.
Algorithm 5.1 (Evaluation of determinant using partial pivoting). This
algorithm nds the value of a determinant of order n n using partial pivoting.
Algorithm Det Partial Pivoting.
//The value of determinant using partial pivoting.//
Let A = [aij ] be an n n matrix.
Step 1. Read the matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. Set k = 1 and sign = 1// sign indicates the sign of the determinant
when interchanges two rows.//
Step 3. Find a pivot from the elements akk , ak+1k , , ank in the kth
That is, |ajk | =
column of A, and let ajk be the pivot.
max{|akk |, |ak+1 k |, . . . , |ank |}.
Step 4. If ajk = 0 then |A| = 0; print the value of |A| and Stop.
Step 5. If j = k then go to Step 6.
Otherwise interchange the kth and jth rows and set sign = sign.

## Solution of System of Linear Equations

283

ajk
times the kth row from the jth row for j = k+1, k+2, . . . , n
akk
ajk
.Rk , where
i.e., for j = k + 1, k + 2, . . . , n do the following Rj = Rj
akk

Rj , Rj are the old and new jth rows respectively.
// This step makes ak+1k , ak+2k , , ank zero.//
Step 7. Increment k by 1 i.e., set k = k+1. If k < n then goto Step 3. Otherwise,
//Triangularization is complete.//
Compute |A| = sign product of diagonal elements.
Print |A| and Stop.
end Det Partial Pivoting.
Step 6. Subtract

Program 5.1
.
/* Program Partial Pivoting
Program to find the value of a determinant using partial pivoting */
#include<stdio.h>
#include<stdlib.h>
#include<math.h>
void main()
{
int n,k,i,j,sign=1;
float a,b,prod,temp;
int max1(float b[],int k, int n);
printf("\nEnter the size of the determinant ");
scanf("%d",&n);
printf("Enter the elements rowwise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++)
scanf("%f",&a[i][j]);
for(k=1;k<=n;k++)
{
for(i=k;i<=n;i++) b[i]=a[i][k];
/* copy from a[k][k] to a[n][k] into b */
j=max1(b,k,n); /* finds pivot position */
if(a[j][k]==0)
{
printf("The value of determinant is 0");
exit(0);
}
if(j!=k) /* interchange k and j rows */
{
sign=-sign;

## 284 Numerical Analysis

for(i=1;i<=n;i++){
temp=a[j][i]; a[j][i]=a[k][i]; a[k][i]=temp;
}
}
for(j=k+1;j<=n;j++) /* makes a[k+1][k] to a[n][k] zero */
{
temp=a[j][k]/a[k][k];
for(i=1;i<=n;i++)
a[j][i]-=temp*a[k][i];
}
} /* end of k loop */
prod=sign;
/* product of diagonal elements */
for(i=1;i<=n;i++) prod*=a[i][i];
printf("The value of the determinant is %f ",prod);
}/* main */
/* finds position of maximum element among n numbers */
int max1(float b[],int k, int n)
{
float temp; int i,j;
temp=fabs(b[k]); j=k; /* initial maximum */
for(i=k+1;i<=n;i++)
if(temp<fabs(b[i])) {temp=fabs(b[i]); j=i;}
return j;
}
A sample of input/output:
Enter the size of the determinant 3
Enter the elements rowwise
0 2 5
1 3 -8
6 5 1
The value of the determinant is -163.000000
Algorithm 5.2 (Evaluation of determinant using complete pivoting). This
algorithm nds the value of a determinant of order n n using complete pivoting.
Algorithm Det Complete Pivoting.
Let A = [aij ] be an n n matrix.
Step 1. Read the matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. Set k = 1 and sign = 1.

## Solution of System of Linear Equations

285

Step 3. Find
a pivot from the elements of the trailing sub-matrix

## akk akk+1 akn

ak+1k ak+1k+1 ak+1n
of A. Let apq be the pivot.

ank ank+1 ann
i.e., |apq | = max{|akk |, . . . , |akn |; |ak+1k |, . . . , |ak+1n |; |ank |, . . . , |ann |}.
Step 4. If apq = 0 then |A| = 0; print the value of |A| and Stop.
Step 5. If p = k then goto Step 6. Otherwise, interchange the kth and the pth
rows and set sign = sign.
Step 6. If q = k then goto Step 7. Otherwise, interchange the kth and the qth
columns and set sign = sign.
ajk
times the kth row to the jth row for j = k + 1, k + 2, . . . , n
Step 7. Subtract
akk
i.e., for j = k + 1, k + 2, . . . , n do the following
ajk
.Rk , where Rj , Rj are the old and new jth rows respecRj = Rj
akk
tively.
Step 8. Increment k by 1 i.e., set k = k + 1.
If k < n then goto Step 3. Otherwise,
compute |A| = sign product of diagonal elements.
Print |A| and Stop.
end Det Complete Pivoting.
Program 5.2
.
/*Program Complete Pivoting
Program to find the value of a determinant using complete pivoting */
#include<stdio.h>
#include<stdlib.h>
#include<math.h>
void main()
{
int n,k,i,j,sign=1,p,q;
float a,prod,max,temp;
printf("\nEnter the size of the determinant ");
scanf("%d",&n);
printf("Enter the elements rowwise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
for(k=1;k<=n;k++)
{
/* finds the position of the pivot element */
max=fabs(a[k][k]); p=k; q=k; /* set initial maximum */

## 286 Numerical Analysis

for(i=k;i<=n;i++)
for(j=k;j<=n;j++)
if(max<fabs(a[i][j])) { max=fabs(a[i][j]); p=i; q=j;}
if(a[p][q]==0)
{
printf("The value of determinant is 0");
exit(0);
}
if(p!=k) /* interchange k and p rows */
{
sign=-sign;
for(i=1;i<=n;i++)
{
temp=a[p][i]; a[p][i]=a[k][i]; a[k][i]=temp;
}
}
if(q!=k) /* interchange k and q columns */
{
sign=-sign;
for(i=1;i<=n;i++)
{
temp=a[i][q]; a[i][q]=a[i][k]; a[i][k]=temp;
}
}
for(j=k+1;j<=n;j++) /* makes a[k+1][k] to a[n][k] zero */
{
temp=a[j][k]/a[k][k];
for(i=1;i<=n;i++)
a[j][i]-=temp*a[k][i];
}
} /* end of k loop */
prod=sign;
for(i=1;i<=n;i++) /* product of diagonal elements*/
prod*=a[i][i];
printf("The value of the determinant is %f ",prod);
}/* main */
A sample of input/output:
Enter the size of the determinant 4
Enter the elements rowwise

## Solution of System of Linear Equations

287

-2 3 8 4
6 1 0 5
-8 3 1 2
3 8 7 10
The value of the determinant is -1273.000000
The general disadvantages of the pivoting is that the symmetry or regularity of the
original matrix may be lost. Partial pivoting requires less time in terms of interchanges
and search for the pivot than the complete pivoting. A combination of partial and
complete pivoting is expected to be very eective not only for computing a determinant
but also for solving system of linear equations. The pivoting brings in stability where a
method becomes unstable for a problem. The pivoting reduces the error due to the loss
of signicant digits.

5.3

Inverse of a Matrix

From the theory of matrices, it is well known that every square non-singular matrix has
unique inverse. The inverse of a matrix A is dened by
A1 =

.
|A|

A11
A12

A1n

(5.7)

and dened as

A21 An1
A22 An2
,

A2n Ann

## where Aij being the cofactor of aij in |A|.

The main diculty of this method is to compute the inverse of the matrix A. From
the denition of adj A it is easy to observe that to compute the matrix adj A, we have
to determine n2 determinants each of order (n 1). So, it is very much time consuming.
Many ecient methods are available to nd the inverse of a matrix, among them GaussJordan is most popular. In the following Gauss-Jordan method is discussed to nd the
inverse of a square non-singular matrix.
5.3.1

Gauss-Jordan Method

In this method, the given matrix A is augmented with a unit matrix of same size, i.e.,
.
if the order of A is n n then the order of the augmented matrix [A..I] will be n 2n.

## 288 Numerical Analysis

The augmented matrix looks like

a a a1n
11 12

.
a a a2n
[A..I] = 21 22

## an1 an2 ann

..
.
..
.
..
.
..
.

0 0

1 0
.

(5.8)

0 1

Then the inverse of A is computed in two stages. In the rst stage, A is converted
into an upper triangular form, using only elementary row operations (Gauss elimination
method discussed in Section 5.5). In the second stage, the upper triangular matrix
(obtained in rst stage) is reduced to an identity matrix by row operations. All these
.
operations are operated on the augmented matrix [A..I]. After completion of these
.
.
stages, the augmented matrix [A..I] is turned to [I..A1 ], i.e., the inverse of A is obtained
from the right half of augmented matrix.
Thus
 .
 ..  Gauss Jordan

I..A1 .
A.I
At the end of the operations the matrix shown in (5.8) reduces to the following form:

.
1 0 0 .. a11 a12 a1n

.. 

 a
0
1

0
a
.
a

21 22
2n .
(5.9)

...

.. 


0 0 1 . a a a
n1

n2

nn

2 4 5
Example 5.3.1 Find the inverse of the following matrix A = 1 1 2 .
3 4 5
.
Solution. The augmented matrix [A..I] can be written as

.
4 5 .. 1 0 0

.
..
.
[A..I] =
1
1
2
.
0
1
0

..
3 4 5 . 0 0 1
2

(5.10)

## Stage I. (Reduction to upper triangular form):

In the rst column 3 is the largest element, thus interchanging rst (R1 ) and third
(R3 ) rows to bring the pivot element 3 to the a11 position. Then (5.10) becomes

## Solution of System of Linear Equations

289

.
4 5 .. 0 0 1

..

.
1
1
2
.
0
1
0

..
2 4 5 . 1 0 0

.
1 4/3 5/3 .. 0 0 1/3


.
1

1 1 2 .. 0 1 0 R1 = 3 R1
.
2 4 5 .. 1 0 0

.
1 4/3 5/3 .. 0 0 1/3


..
R = R2 R1 ; R = R3 2R1

0
7/3
1/3
.
0
1
1/3
3

2
..
0 4/3 5/3 . 1 0 2/3
3

7
(The largest element (in magnitude) in the second column is , which is at the a22
3
position and so there is no need to interchange any rows).

.
1 4/3 5/3 .. 0 0
1/3


.
3
4 


## 0 1 1/7 .. 0 3/7 1/7 R2 = 7 R2 ; R3 = R3 3 R2

.
0 0 13/7 .. 1 4/7 6/7

.
1 4/3 5/3 .. 0
0
1/3


.
7

## 0 1 1/7 .. 0 3/7 1/7 R3 = 13 R3

.
0 0
1 .. 7/13 4/13 6/13
Stage II. (Make the left half a unit matrix):

.
1 0 13/7 .. 0
4/7 1/7


.
4

## 0 1 1/7 .. 0 3/7 1/7 R1 = R1 3 R2

.
0 0 1 .. 7/13 4/13 6/13

.
1 0 0 .. 1
0
1


.
13
1


## 0 1 0 .. 1/13 5/13 1/13 R1 = R1 7 R3 ; R2 = R2 + 7 R3

.
0 0 1 .. 7/13 4/13 6/13
The left hand becomes a unit matrix, thus the inverse of the given matrix is

1
0
1
1/13 5/13 1/13 .
7/13 4/13 6/13

## 290 Numerical Analysis

Algorithm 5.3 (Matrix inverse). The following algorithm computes the inverse
of a non-singular square matrix of order n n and if the matrix is singular it prints
the message the matrix is singular and hence not invertible.
Algorithm Matrix Inversion (using partial pivoting).
Let A = [aij ] be an n n matrix.
Step 1. Read the matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. //Augment the matrix A.//
Augment the matrix A by a unit matrix of order n n. The resultant
matrix A becomes of order n 2n.

0, for i = j
1, for i = j
for i, j = 1, 2, . . . , n.

i.e., ai n+j =

## Stage I. Make upper triangular form.

Step 3. Set k = 1.
Step 4. Find a pivot from the elements akk , ak+1 k , . . . , ank in the kth column of
A and let ajk be the pivot.
Step 5. If ajk = 0 then print the matrix is singular and hence not invertible
and Stop.
Step 6. If j = k then goto Step 7.
Otherwise interchange the kth and jth rows.
Step 7. If akk = 1 then divide all the elements of kth row by akk .
Subtract ajk times the kth row to the jth row for
j = k + 1, k + 2, , 2n;
i.e., Rj = Rj ajk Rk .
//This step makes ak+1 k , ak+2 k , . . . , ank zero.//
Step 8. Increase k by 1 i.e., set k = k + 1.
If k < n then goto Step 4. Otherwise goto Step 9.
//Stage I is completed.//
Stage II. //Make the left half of A a unit matrix.//
Step 9. Set k = 2.
Step 10. Subtract ajk times the kth row to the jth row for
j = k 1, k 2, . . . , 1.
Step 11. Increase k by 1.
If k < n then goto Step 10.
Otherwise, print the right half of A as inverse of A and Stop.
end Matrix Inversion

## Solution of System of Linear Equations

Program 5.3
.
/* Program Matrix Inverse
Program to find the inverse of a square matrix using
partial pivoting */
#include<stdio.h>
#include<stdlib.h>
#include<math.h>
#define zero 0.00001
void main()
{
int n,m,k,i,j;
float a,temp;
printf("\nEnter the size of the matrix ");
scanf("%d",&n);
printf("Enter the elements rowwise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
/* augment the matrix A */
for(i=1;i<=n;i++) for(j=1;j<=n;j++) a[i][n+j]=0;
for(i=1;i<=n;i++) a[i][n+i]=1;
m=2*n;
for(k=1;k<=n;k++)
{
/* finds pivot element and its position */
temp=fabs(a[k][k]); j=k; /* initial maximum */
for(i=k+1;i<=n;i++)
if(temp<fabs(a[i][k])){
temp=fabs(a[i][k]); j=i;
}
if(fabs(a[j][k])<=zero) /* if a[j][k]=0 */
{
printf("The matrix is singular and is not invertible");
exit(0);
}
if(j!=k) /* interchange k and j rows */
{
for(i=1;i<=m;i++){
temp=a[j][i]; a[j][i]=a[k][i]; a[k][i]=temp;
}
}

291

## 292 Numerical Analysis

if(a[k][k]!=1)
{
temp=a[k][k];
for(i=1;i<=m;i++) a[k][i]/=temp;
}
for(j=k+1;j<=n;j++) /* makes a[k+1][k] to a[n][k] zero */
{
temp=a[j][k];
for(i=1;i<=m;i++)
a[j][i]-=temp*a[k][i];
}
} /* end of k loop */
/* make left half of A to a unit matrix */
for(k=2;k<=n;k++)
{
for(j=k-1;j>=1;j--)
{
temp=a[j][k];
for(i=1;i<=m;i++) a[j][i]-=temp*a[k][i];
}
}
printf("\nThe inverse matrix is \n");
for(i=1;i<=n;i++)
{
for(j=n+1;j<=m;j++)
printf("%f ",a[i][j]); printf("\n");
}
}/* main */
A sample of input/output:
Enter the size of the matrix 3
Enter the elements rowwise
0 1 2
3 -2 1
4 3 2
The inverse matrix is
-0.218750 0.125000 0.156250
-0.062500 -0.250000 0.187500
0.531250 0.125000 -0.093750

293

## Complexity of the algorithm

Step 4 determines the maximum among nk+1 elements and takes nk+1 comparisons.
Step 6 takes O(n) operations if the kth and jth rows need to interchange. For a xed j,
Step 7 needs n k additions and n k multiplications. Since j is running from k + 1 to
n, so the total time taken by Step 7 is (n k)2 . The Step 4 to Step 7 are repeated for n
n

(n k)2 + O(n) + n k + 1 = O(n3 )
times (k = 1, 2, . . . , n), therefore, Stage I takes
k=1

operations.
Similarly, Stage II takes O(n3 ) time. Hence the time complexity to compute the
inverse of a non-singular matrix is O(n3 ).
Since the Stage I is similar to the algorithm Det Partial Pivoting, so the time
complexity to compute the determinant is O(n3 ).

5.4

## Matrix Inverse Method

The system of equations (5.1) can be written in the matrix form (5.3) as
Ax = b
where A, b and x are dened in (5.4).
The solution of Ax = b is given
x = A1 b,

(5.11)

## where A1 is the inverse of the matrix A.

Once the inverse of A is known then post multiplication of it with b gives the solution
vector x.
Example 5.4.1 Solve the following system of equations by matrix inverse method
x + 2y + 3z = 10, x + 3y 2z = 7, 2x y + z = 5.
Solution. The given system of equations is Ax = b, where

1 2 3
x
10
A = 1 3 2 ,
x = y ,
b = 7 .
2 1 1
z
5


1 2 3


Now, |A| =  1 3 2  = 30 = 0.
 2 1 1 
That is, A is non-singular and hence A1 exists.

## 294 Numerical Analysis

1 5 13
5.
7 5
1

1 5 13
1
5 5
5.
=
Thus, A1 =
|A|
30 7 5
1

1 5 13
90
10
3
1
1
1

5 5
5
60 = 2 .
7 =
Therefore, x = A b =
30 7 5
30 30
1
5
1
Hence the required solution is x = 3, y = 2, z = 1.
Algorithm 5.4 (Matrix inverse method). This algorithm is used to solve a
system of linear equations Ax = b, where A = [aij ]nn , x = (x1 , x2 , . . . , xn )t ,
b = (b1 , b2 , . . . , bn )t , by matrix inverse method.
Algorithm Matrix Inverse Method
Step 1. Read the coecient matrix A = [aij ], i, j = 1, 2, . . . , n and the right
hand vector b = (b1 , b2 , . . . , bn )t .
Step 2. Compute the inverse of A by the algorithm Matrix Inverse.
Step 3. If A is invertible then compute x = A1 b and print x =
(x1 , x2 , . . . , xn )t .
Otherwise, print A is singular and the system has either no solution or
has innitely many solutions.
end Matrix Inverse Method
Program 5.4
.
/* Program Matrix Inverse Method
Program to find the solution of a system of linear
equation by matrix inverse method. Partial pivoting
is used to find matrix inverse. */
#include<stdio.h>
#include<stdlib.h>
#include<math.h>
#define zero 0.00001
float a,ai;
int n;
int matinv();
void main()
{
int i,j;
float b,x;

## printf("\nEnter the size of the coefficient matrix ");

scanf("%d",&n);
printf("Enter the elements rowwise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("Enter the right hand vector ");
for(i=1;i<=n;i++) scanf("%f",&b[i]);
i=matinv();
if(i==0)
{
printf("Coefficient matrix is singular: ");
printf("The system has either no solution or many solutions");
exit(0);
}
for(i=1;i<=n;i++)
{
x[i]=0;
for(j=1;j<=n;j++) x[i]+=ai[i][j]*b[j];
}
printf("Solution of the system is\n ");
for(i=1;i<=n;i++) printf("%8.5f ",x[i]);
} /* main */
/* function to find matrix inverse */
int matinv()
{
int i,j,m,k; float temp;
/* augment the matrix A */
for(i=1;i<=n;i++) for(j=1;j<=n;j++) a[i][n+j]=0;
for(i=1;i<=n;i++) a[i][n+i]=1;
m=2*n;
for(k=1;k<=n;k++)
{
/* finds pivot element and its position */
temp=fabs(a[k][k]); j=k; /* initial maximum */
for(i=k+1;i<=n;i++)
if(temp<fabs(a[i][k]))
{
temp=fabs(a[i][k]); j=i;
}

295

## 296 Numerical Analysis

if(fabs(a[j][k])<=zero) /* if a[j][k]=0 */
{
printf("The matrix is singular and is not invertible");
return(0);
}
if(j!=k) /* interchange k and j rows */
{
for(i=1;i<=m;i++)
{
temp=a[j][i]; a[j][i]=a[k][i]; a[k][i]=temp;
}
}
if(a[k][k]!=1)
{
temp=a[k][k];
for(i=1;i<=m;i++) a[k][i]/=temp;
}
for(j=k+1;j<=n;j++) /* makes a[k+1][k] to a[n][k] zero */
{
temp=a[j][k];
for(i=1;i<=m;i++)
a[j][i]-=temp*a[k][i];
}
} /* end of k loop */
/* make left half of A to a unit matrix */
for(k=2;k<=n;k++)
for(j=k-1;j>=1;j--){
temp=a[j][k];
for(i=1;i<=m;i++) a[j][i]-=temp*a[k][i];
}
printf("\nThe inverse matrix is \n");
for(i=1;i<=n;i++)
for(j=n+1;j<=m;j++) ai[i][j-n]=a[i][j];
return(1);
}/* matinv */
A sample of input/output:
Enter the size of the matrix 3
Enter the elements rowwise

## Solution of System of Linear Equations

297

0 1 2
3 -2 1
4 3 2
The inverse matrix is
-0.218750 0.125000 0.156250
-0.062500 -0.250000 0.187500
0.531250 0.125000 -0.093750
Complexity
The time complexity of this method is O(n3 ) as the method involves computation of
A1 .

5.5

## In this method, the variables are eliminated by a process of systematic elimination.

Suppose the system has n variables and n equations of the form (5.1). This procedure
reduces the system of linear equations to an equivalent upper triangular system which
can be solved by backsubstitution. To convert an upper triangular system, x1 is
eliminated from second equation to nth equation, x2 is eliminated from third equation
to nth equation, x3 is eliminated from fourth equation to nth equation, and so on and
nally, xn1 is eliminated from nth equation.
To eliminate x1 , from second, third, , and nth equations the rst equation is mula31
an1
a21
,
, ...,
respectively and successively added with the second,
tiplied by
a11
a11
a11
third, , nth equations (assuming that a11 = 0). This gives
a11 x1 + a12 x2 + a13 x3 + + a1n xn = b1
(1)

(1)

(1)

(1)

(1)

(1)

(1)

(1)

## a22 x2 + a23 x3 + + a2n xn = b2

a32 x2 + a33 x3 + + a3n xn = b3

(5.12)

(1)
an2 x2

where
(1)

(1)

(1)
+ an3 x3 + + a(1)
nn xn = bn ,

aij = aij

ai1
a1j ; i, j = 2, 3, . . . , n.
a11

Again, to eliminate x2 from the third, forth, . . ., and nth equations the second equa(1)
(1)
(1)
a
a42
an2
(1)
tion is multiplied by 32
,

,
.
.
.,

(1)
(1)
(1)
a22
a22
a22

## 298 Numerical Analysis

successively added to the third, fourth, . . ., and nth equations to get the new system of
equations as
a11 x1 + a12 x2 + a13 x3 + + a1n xn = b1
(1)

(1)

(1)

(1)

(2)

(2)

(2)

## a22 x2 + a23 x3 + + a2n xn = b2

a33 x3 + + a3n xn = b3

(5.13)

(2)

(2)
an3 x3 + + a(2)
nn xn = bn ,

where

(1)

(2)

(1)

aij = aij

ai2

(1)
a ;
(1) 2j
a22

i, j = 3, 4, . . . , n.

## Finally, after eliminating xn1 , the above system of equations become

a11 x1 + a12 x2 + a13 x3 + + a1n xn = b1
(1)

(1)

(1)

(1)

(2)

(2)

(2)

## a22 x2 + a23 x3 + + a2n xn = b2

a33 x3 + + a3n xn = b3

(5.14)

a(n1)
xn = b(n1)
,
nn
n
where,
(k1)

(k)

(k1)

aij = aij

aik

(k1)
akk

(k1)

akj

(0)

## i, j = k + 1, . . . , n; k = 1, 2, . . . , n 1, and apq = apq ; p, q = 1, 2, . . . , n.

Now, by back substitution, the values of the variables can be found as follows:
(n1)
bn
From last equation we have, xn = (n1) , from the last but one equation, i.e., (n1)th
ann
equation, one can nd the value of xn1 and so on. Finally, from the rst equation we
obtain the value of x1 .
(k)
The evaluation of the elements aij s is a forward substitution and the determination of the values of the variables xi s is a back substitution since we rst determine
the value of the last variable xn .
Note 5.5.1 The method described above assumes that the diagonal elements are nonzero. If they are zero or nearly zero then the above simple method is not applicable to
solve a linear system though it may have a solution. If any diagonal element is zero or
very small then partial pivoting should be used to get a solution or a better solution.

299

## It is mentioned earlier that if the system is diagonally dominant or real symmetric

and positive denite then no pivoting is necessary.
Example 5.5.1 Solve the equations by Gauss elimination method.
2x1 + x2 + x3 = 4, x1 x2 + 2x3 = 2, 2x1 + 2x2 x3 = 3.
Solution. Multiplying the second and third equations by 2 and 1 respectively and
subtracting them from rst equation we get
2x1 + x2 + x3 = 4
3x2 3x3 = 0
x2 + 2x3 = 1.
Multiplying third equation by 3 and subtracting from second equation we obtain
2x1 + x2 + x3 = 4
3x2 3x3 = 0
3x3 = 3.
From the third equation x3 = 1, from the second equations x2 = x3 = 1 and from
the rst equation 2x1 = 4 x2 x3 = 2 or, x1 = 1.
Therefore the solution is x1 = 1, x2 = 1, x3 = 1.
Example 5.5.2 Solve the following system of equations by Gauss elimination
method (use partial pivoting).
x2 + 2x3 = 5
x1 + 2x2 + 4x3 = 11
3x1 + x2 5x3 = 12.
Solution. The largest element (the pivot) in the coecients of the variable x1 is 3,
attained at the third equation. So we interchange rst and third equations
3x1 + x2 5x3 = 12
x1 + 2x2 + 4x3 = 11
x2 + 2x3 = 5.
Multiplying the second equation by 3 and adding with the rst equation we get,
3x1 + x2 5x3 = 12
x2 + x3 = 3
x2 + 2x3 = 5

## 300 Numerical Analysis

The second pivot is 1, which is at the positions a22 and a32 . Taking a22 = 1 as
pivot to avoid interchange of rows. Now, subtracting the third equation from second
equation, we obtain
3x1 + x2 5x3 = 12
x2 + x3 = 3
x3 = 2.
Now by back substitution, the values of x3 , x2 , x1 are obtained as
1
x3 = 2, x2 = 3 x3 = 1, x1 = (12 x2 + 5x3 ) = 1.
3
Hence the solution is x1 = 1, x2 = 1, x3 = 2.
Algorithm 5.5 (Gauss elimination). This algorithm solves a system of equations
Ax = b, where A =[aij ]nn , x = (x1 , x2 , . . . , xn )t , b = (b1 , b2 , . . . , bn )t , by Gauss
elimination method.
Algorithm Gauss Elimination (using partial pivoting)
Step 1. Read the matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. Set ain+1 = bi for i = 1, 2, . . . , n.
Then the augmented matrix A becomes of order n (n + 1).
Step 3. Set k = 1.
Step 4. Find the pivot from the elements akk , ak+1k , . . . , ank . Let ajk be the
pivot.
Step 5. If ajk = 0 then // A is singular.//
Print the system has either no solution or innite many solutions, and
Stop.
Step 6. If j = k then interchange the rows j and k.
Step 7. Subtract ajk /akk times the kth row to the jth row for j = k + 1, k +
2, . . . , n.
Step 8. Increase k by 1.
If k = n then //forward substitution is over.//
go to Step 9.
Otherwise go to Step 4.
Step 9. //Back substitution.//
xn = an n+1 /ann .
Compute xi s using the expression


n

1
ai n+1
aij xj , for i = n 1, n 2, . . . , 1.
xi =
aii
j=i+1

## Step 10. Print x =(x1 , x2 , . . . , xn )t as solution.

end Gauss Elimination

## Solution of System of Linear Equations

Program 5.5
.
/* Program Gauss-elimination
Program to find the solution of a system of linear equations by
Gauss elimination method. Partial pivoting is used. */
#include<stdio.h>
#include<stdlib.h>
#include<math.h>
#define zero 0.00001
void main()
{
int i,j,k,n,m;
float a,b,x,temp;
printf("\nEnter the size of the coefficient matrix ");
scanf("%d",&n);
printf("Enter the elements rowwise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("Enter the right hand vector ");
for(i=1;i<=n;i++) scanf("%f",&b[i]);
/* augment A with b[i], i.e., a[i][n+1]=b[i] */
m=n+1;
for(i=1;i<=n;i++) a[i][m]=b[i];
for(k=1;k<=n;k++)
{
/* finds pivot element and its position */
temp=fabs(a[k][k]); j=k; /* initial maximum */
for(i=k+1;i<=n;i++)
if(temp<fabs(a[i][k]))
{
temp=fabs(a[i][k]); j=i;
}
if(fabs(a[j][k])<=zero) /* if a[j][k]=0 */
{
printf("The matrix is singular:");
printf("The system has either no solution or many solutions");
exit(0);
}
if(j!=k) /* interchange k and j rows */
{
for(i=1;i<=m;i++)

301

## 302 Numerical Analysis

{
temp=a[j][i]; a[j][i]=a[k][i]; a[k][i]=temp;
}
}
for(j=k+1;j<=n;j++) /* makes a[k+1][k] to a[n][k] zero */
{
temp=a[j][k]/a[k][k];
for(i=1;i<=m;i++)
a[j][i]-=temp*a[k][i];
}
} /* end of k loop */
/* forward substitution is over */
/* backward substitution */
x[n]=a[n][m]/a[n][n];
for(i=n-1;i>=1;i--)
{
x[i]=a[i][m];
for(j=i+1;j<=n;j++) x[i]-=a[i][j]*x[j];
x[i]/=a[i][i];
}
printf("Solution of the system is\n ");
for(i=1;i<=n;i++) printf("%8.5f ",x[i]);
} /* main */
A sample of input/output:
Enter the size of the coefficient matrix 3
Enter the elements rowwise
1 1 1
2 -1 3
3 1 -1
Enter the right hand vector
3 16 -3
Solution of the system is
1.00000 -2.00000 4.00000

5.6

## In Gauss elimination method, the coecient matrix is reduced to an upper triangular

form and the solution is obtained by back substitution. But, the Gauss-Jordan method
reduces the coecient matrix to a diagonal matrix rather than upper triangular matrix

## Solution of System of Linear Equations

303

and produces the solution of the system without using the back substitution. At the end
of Gauss-Jordan method the system of equations (5.2) reduces to the following form:

1
0

0
1

x1
x2
..
.
xn

b1
b
2
= .. .
.

(5.15)

bn

## The solution of the system is given by

x1 = b1 , x2 = b2 , . . . , xn = bn .
Thus the Gauss-Jordan method gives
 . 
Gauss Jordan
I..b .

 .. 
A.b

(5.16)

## Generally, the Gauss-Jordan method is not used to solve a system of equations,

because it is more costly than the Gauss elimination method. But, this method is used
to nd the matrix inverse (discussed in Section 5.3).
Example 5.6.1 Solve the following equations by Gauss-Jordan elimination method.
x1 + x2 + x3 = 3
2x1 + 3x2 + x3 = 6
x1 x2 x3 = 3.

1 1 1
3
x1
Solution. Here A = 2 3 1 , x = x2 and b = 6 .
x3
1 1 1
3
 .. 
The augmented matrix A.b is

 .. 
A.b =
2

1
3

1 1

1
1

0 2

.
1 .. 3

1 .. 6
.
1 .. 3

.
1 .. 3
R = R2 2R1 ,
.
2
1 .. 0 R = R3 R1
3
.
2 .. 6

## 304 Numerical Analysis

1 1

0 1

0 0

1 1

0 1

0 0

1 0

0 1

0 0

1 0

0 1
0 0

.
1 .. 3

.

1 .. 0 R3 = R3 + 2R2
.
4 .. 6

.
1 .. 3

1
.

1 .. 0 R3 = 4 R3
.
1 .. 3/2

.
2 .. 3

.

1 .. 0 R1 = R1 R2
.
1 .. 3/2

.
0 .. 0
R = R1 2R3 ,
.
1
0 .. 3/2 R = R2 + R3
2
.
1 .. 3/2

## The equivalent system of equations are

=0
x1
= 3/2
x2
x3 = 3/2.
Hence the required solution is x1 = 0, x2 = 3/2, x3 = 3/2.

5.7
5.7.1

## Method of Matrix Factorization

LU Decomposition Method

## This method is also known as factorization or LU decomposition method or Crouts

reduction method.
Let the system of linear equations be
Ax = b

(5.17)

## where A, x, b are given by (5.4).

The matrix A can be factorized into the form A = LU, where L and U are the
lower and upper triangular matrices respectively. If the principal minors of A are nonsingular, i.e.,




 a11 a12 a13 
 a11 a12 


 = 0,  a21 a22 a23  = 0, , |A| = 0
a11 = 0, 
(5.18)



a21 a22
 a31 a32 a33 

## then this factorization is possible and it is unique.

The matrices L and U are of the form

l11 0 0 0
u11 u12
l21 l22 0 0
0 u22

.. .. .. .. ..
..
..
. . . . .
.
.
0 0
ln1 ln2 ln3 lnn

u1n
u2n

u3n
.
..
..
.
.
0 unn

u13
u23
u33
..
.

305

(5.19)

## The equation Ax = b becomes LUx = b. Let Ux = z then Lz = b, where

z = (z1 , z2 , . . . , zn )t is an intermediate variable vector. The value of z i.e., z1 , z2 , . . . , zn
can be determined by forward substitution in the following equations.
l11 z1
l21 z1 + l22 z2
l31 z1 + l32 z2 + l33 z3

## ln1 z1 + ln2 z2 + ln3 z3 + + lnn zn

=
=
=

b1
b2
b3

bn .

(5.20)

After determination of z, one can compute the value of x i.e., x1 , x2 , . . . , xn from the
equation Ux = z i.e., from the following equations by the backward substitution.
u11 x1 + u12 x2 + u13 x3 + + u1n xn
u22 x2 + u23 x3 + z2n xn
u33 x3 + u23 x3 + u3n xn

## un1n1 xn1 + un1n xn

unn xn

=
=
=

=
=

z1
z2
z3

zn1
zn .

(5.21)

When uii = 1, for i = 1, 2, . . . , n, then the method is known as Crouts decomposition method. When lii = 1, for i = 1, 2, . . . , n then the method is known as
Doolittles method for decomposition. In particular, when lii = uii for i = 1, 2, . . . , n
then the corresponding method is called Choleskys decomposition method.
Procedure to compute L and U
Here,
we assume that uii = 1 for i = 1, 2, . . . , n. From the relation LU =A, i.e., from
l11 u13
l11 u1n
l11 l11 u12

## l21 l21 u12 + l22 l21 u13 + l22 u23

l21 u1n + l22 u2n

l31 l31 u12 + l32 l31 u13 + l32 u23 + l33 l31 u1n + l32 + u2n + l33 u3n

..
..
..
..
..

.
.
.
.
.
ln1 ln1 u12 + ln2 ln1 u13 + ln2 u23 + ln3 ln1 u1n + ln2 u2n + + lnn

## a11 a12 a13 a1n

a21 a22 a23 a2n

=

a31 a32 a33 ann
we obtain
a1j
li1 = ai1 , i = 1, 2, . . . , n and u1j =
, j = 2, 3, . . . , n.
l11
The second column of L and the second row of U are determined from the relations
li2 = ai2 li1 u12 , for i = 2, 3, . . . , n,
a2j l21 u1j
for j = 3, 4, . . . , n.
u2j =
l22
Next, third column of L and third row of U are determined in a similar way.
In general, lij and uij are given by
lij = aij
aij
uij =

j1

k=1
i1


lik ukj , i j

(5.22)

lik ukj

k=1

, i<j
lii
uii = 1, lij = 0, j > i and uij = 0, i > j.

(5.23)

## Alternatively, the vectors z and x can be determined from the equations

z = L1 b
and

x=U

z.

(5.24)
(5.25)

It may be noted that the computation of inverse of a triangular matrix is easier than
an arbitrary matrix.
The inverse of A can also be determined from the relation
A1 = U1 L1 .

(5.26)

## Some properties of triangular matrices

Let L = [lij ] and U = [uij ] denote respectively the lower and upper triangular matrices.
The determinant of a triangular matrix is the product of the diagonal elements
n
n
*
*
lii and |U| =
uii .
i.e., |L| =
i=1

i=1

matrix.

## Solution of System of Linear Equations

307

The inverse of lower (upper) triangular matrix is also a lower (upper) triangular
matrix.
*
 *

n
n
Since A = LU, |A| = |LU| = |L||U| =
lii
uii .
i=1

i=1

## It may be remembered that the computation of determinant by LU decomposition

is not a better method since it may fail or become unstable due to vanishing or nearvanishing leading minors.
Example 5.7.1 Factorize the matrix

2 2 1
A = 5 1 3
3 4 1

into the form LU, where L and U are lower and upper triangular matrices and hence
solve the system of equations 2x1 2x2 +x3 = 2, 5x1 +x2 3x3 = 0, 3x1 +4x2 +x3 =
9. Determine L1 and U1 and hence nd A1 . Also determine |A|.

2 2 1
Solution. Let 5 1 3
3 4 1

l11 u13
1 u12 u13
l11 l11 u12
l11 0 0
.
= l21 l22 0 0 1 u23 = l21 l21 u12 + l22 l21 u13 + l22 u23
0 0 1
l31 l31 u12 + l32 l31 u13 + l32 u23 + l33
l31 l32 l33
Comparing both sides, we have
l31 = 3
l11 = 2, l21 = 5
or, u12 = 2/l11 = 1
l11 u12 = 2
or, u13 = 1/l11 = 1/2
l11 u13 = 1
or, l22 = 1 l21 u12 = 6
l21 u12 + l22 = 1
or, l32 = 4 l31 u12 = 7
l31 u12 + l32 = 4
or, u23 = (3 l21 u13 )/l22 = 11/12
l21 u13 + l22 u23 = 3
l31 u13 + l32 u23 + l33 = 1 or, l33 = 1 l31 u13 l32 u23 = 71/12.
Hence L and U are given by

1
2 0 0
1 1
2
.
L = 5 6 0 ,
U = 0 1 11
12
71
3 7 12
0 0
1
Second Part. The given system
of equations
can be
written asAx = b, where

2
2 2 1
x1
A = 5 1 3 , x = x2 , b = 0 .
x3
9
3 4 1

## 308 Numerical Analysis

Using A = LU, the equation Ax = b reduces to LUx = b. Let Ux = y. Then
Ly = b.
From the relation Ly = b, we have

2 0 0
2
y1
5 6 0 y2 = 0 ,
9
y3
3 7 71
12
That is,
2y1
5y1 + 6y2
3y1 + 7y2 +

= 2,
= 0,
71
y3 = 9.
12

## Solution of this system is y1 = 1, y2 = 56 , y3 = 2.

Thus y = (1, 5/6, 2)t .
Now, from the relation Ux = y we have

1
1
1 1
x1
2
0 1 11 x2 = 5 ,
6
12
x3
2
0 0
1
i.e.,
1
x1 x2 + x3 = 1
2
11
5
x2 x3 =
12
6
x3 = 2.
Solution of this system of equations is
5 11
1
x3 = 2, x2 = +
2 = 1, x1 = 1 + x2 x3 = 1.
6 12
2
Hence the required solution is x1 = 1, x2 = 1, x3 = 2.
Third Part. Applying the Gauss-Jordan method to nd L1 . The necessary augmented matrix is

.
2 0 0 .. 1 0 0

 .. 
.

L.I =
5 6 0 .. 0 1 0
..
3 7 71
12 . 0 0 1

## Solution of System of Linear Equations

309

.
1 0 0 .. 12 0 0


.. 5
R1 = 1 R1 , R2 = R2 5R1 , R3 = R3 3R1

0
6
0
.

1
0

2
2
71 ..
3
0 7 12 . 2 0 1

.
1
1 0 0 ..
0
0
2


1
.. 5



1

0 1 0 . 12 6 0 R2 = 6 R2 , R3 = R3 7R2
.. 17 7
0 0 71
12 .
12 6 1

.
1
0
0
1 0 0 ..
2

.. 5
.
1

0
0 1 0 . 12

6
.. 17 14 12
0 0 1 . 71 71 71

1/2
0
0
0 .
Hence L1 = 5/12 1/6
17/71 14/71 12/71
The value of U1 can also be determined in a similar way. Here we apply another
method based on the property inverse of a triangular matrix is a triangular matrix
of same shape.

1 b12 b13
Let U1 = 0 1 b23 .
0 0 1
1
Then U U = I gives

1 b12 b13
1 1 1/2
1 0 0
0 1 b23 0 1 11/12 = 0 1 0 .
0 0 1
0 0
1
0 0 1

1 0 0
1 1 + b12 12 11
12 b12 + b13
= 0 1 0.
0
i.e.,
1
11
12 + b23
0 0 1
0
0
1
Equating both sides
1 11
1 11
5
1 + b12 = 0 or, b12 = 1, b12 + b13 = 0 or, b13 = + b12 =
2 12
2 12
12
11
11
+ b23 = 0 or, b23 = .
12
12
Hence

1 1 5/12
U1 = 0 1 11/12 .
0 0 1

## 310 Numerical Analysis

Therefore,

A1

1 1 5/12
1/2
0
0
0
= U1 L1 = 0 1 11/12 5/12 1/6
0 0 1
17/71 14/71 12/71

## 13/71 6/71 5/71

= 14/71 1/71 11/71 .
17/71 14/71 12/71


71 
1 = 71.
Last Part. |A| = |L||U| = 2 6
12
Algorithm 5.6 (LU decomposition). This algorithm nds the solution of a system of linear equations using LU decomposition method. Assume that the principal
minors of all order are non-zero.
Algorithm LU-decomposition
Let Ax = b be the systems of equations and A = [aij ], b = (b1 , b2 , . . . , bn )t ,
x = (x1 , x2 , . . . , xn )t .
//Assume that the principal minors of all order are non-zero.//
//Determine the matrices L and U.//
Step 1. Read the matrix A = [aij ], i, j = 1, 2, . . . , n and the right
hand vector b = (b1 , b2 , . . . , bn )t .
1j
for j = 2, 3, . . . , n;
Step 2. li1 = ai1 for i = 1, 2, . . . , n; u1j = al11
uii = 1 for i = 1, 2, . . . , n.
Step 3. For i, j = 2, 3, . . . , n compute the following
j1

lik ukj , i j
lij = aij


k=1
i1


uij = aij


lik ukj /lii , i < j.

k=1

## Step 4. //Solve the system Lz = b by forward substitution.//

i1


b1
1
bi
, zi =
lij zj for i = 2, 3, . . . , n.
z1 =
l11
lii
j=1

## Step 5. //Solve the system Ux = z by backward substitution.//

Set xn = zn ;
n

xi = zi
uij xj for i = n 1, n 2, . . . , 1.
j=i+1

Print x1 , x2 , . . . , xn as solution.
end LU-decomposition

## Solution of System of Linear Equations

Program 5.6
.
/* Program LU-decomposition
Solution of a system of equations by LU decomposition method.
Assume that all order principal minors are non-zero. */
#include<stdio.h>
void main()
{
float a,l,u,z,x,b;
int i,j,k,n;
printf("\nEnter the size of the coefficient matrix ");
scanf("%d",&n);
printf("Enter the elements rowwise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("Enter the right hand vector ");
for(i=1;i<=n;i++) scanf("%f",&b[i]);
/* computations of L and U matrices */
for(i=1;i<=n;i++) l[i]=a[i];
for(j=2;j<=n;j++) u[j]=a[j]/l;
for(i=1;i<=n;i++) u[i][i]=1;
for(i=2;i<=n;i++)
for(j=2;j<=n;j++)
if(i>=j)
{
l[i][j]=a[i][j];
for(k=1;k<=j-1;k++) l[i][j]-=l[i][k]*u[k][j];
}
else
{
u[i][j]=a[i][j];
for(k=1;k<=i-1;k++) u[i][j]-=l[i][k]*u[k][j];
u[i][j]/=l[i][i];
}
printf("\nThe lower triangular matrix L\n");
for(i=1;i<=n;i++)
{
for(j=1;j<=i;j++) printf("%f ",l[i][j]);
printf("\n");
}
printf("\nThe upper triangular matrix U\n");

311

## 312 Numerical Analysis

for(i=1;i<=n;i++)
{
for(j=1;j<i;j++) printf("
");
for(j=i;j<=n;j++) printf("%f ",u[i][j]);
printf("\n");
}
/* solve Lz=b by forward substitution */
z=b/l;
for(i=2;i<=n;i++)
{
z[i]=b[i];
for(j=1;j<=i-1;j++) z[i]-=l[i][j]*z[j];
z[i]/=l[i][i];
}
/* solve Ux=z by backward substitution */
x[n]=z[n];
for(i=n-1;i>=1;i--)
{
x[i]=z[i];
for(j=i+1;j<=n;j++) x[i]-=u[i][j]*x[j];
}
printf("The solution is ");
for(i=1;i<=n;i++) printf("%f ",x[i]);
} /* main */
A sample of input/output:
Enter the size of the coefficient matrix 3
Enter the elements rowwise
4 2 1
2 5 -2
1 -2 7
Enter the right hand vector
3 4 5
The lower triangular matrix L
4.000000
2.000000 4.000000
1.000000 -2.500000 5.187500

313

## The upper triangular matrix U

1.000000 0.500000 0.250000
1.000000 -0.625000
1.000000
The solution is -0.192771 1.325301 1.120482

5.8

## Gauss Elimination Method to the Find Inverse of a Matrix

 . 
Conventionally, the Gauss elimination is applied to the augmented matrix A..b . This
 . 
method can also be applied to augmented matrix A..I . In this method, the matrix
A(= LU) becomes an upper triangular matrix U and the unit matrix I becomes the
lower triangular matrix, which is the inverse of L. Then the relation AA1 = I becomes
LUA1 = I i.e.,
UA1 = L1 .

(5.27)

The left hand side of (5.27) is a lower triangular matrix and also the matrices U and
L1 are known. Hence by back substitution the matrix A1 can be determined easily.
This method is illustrated by an example below.
Example
5.8.1 Find
the inverse of the matrix A using Gauss elimination method
1 3 4
where A = 1 0 2 .
2 3 1
 . 
Solution. The augmented matrix A..I is

.
1 3 4 .. 1 0 0

 .. 
.

A.I =
1 0 2 .. 0 1 0
.
2 3 1 .. 0 0 1

1 3 4

R2 R2 R1

0 3 2
R3 R3 + 2R1
0 9 9

..
. 1 0 0

..

. 1 1 0
..
. 2 0 1

..
4
.
1
0
0

R3 R3 + 3R2
..

0
3
2
.
1
1
0

..
0 0 3 . 1 3 1
1

## 314 Numerical Analysis

1 3 4
1
1

Here U = 0 3 2 , L = 1
0 0 3
1

## x11 x12 x13

Let A1 = x21 x22 x23 .
x31 x32 x33
Since, UA1 = L1 ,

1 3 4
x11
0 3 2 x21
x31
0 0 3

0 0
1 0.
3 1

x12 x13
1 0 0
x22 x23 = 1 1 0 .
x32 x33
1 3 1

This implies
x11 + 3x21 + 4x31 = 1
3x21 2x31 = 1
3x31 = 1

1
5
2
These equations give x31 = , x21 = , x11 = .
3
9
3
Again, x12 + 3x22 + 4x32 = 0
3x22 2x32 = 1
3x32 = 3
Solution of this system is x32 = 1, x22 = 1, x12 = 1
x13 + 3x23 + 4x33 = 0
3x23 2x33 = 0
3x33 = 1
1
2
2
Then, x33 = , x23 = , x13 = is the solution of the above equations.
3
9
3
2/3 1 2/3
Hence A1 = 5/9 1 2/9 .
1/3 1 1/3

5.9

Cholesky Method

If the coecient matrix A is symmetric and positive denite then this method is applicable to solve the system Ax = b. This method is also known as square-root method.
Since A is symmetric then A can be written as
A = LLt ,

(5.28)

## where L = [lij ], lij = 0, i < j, a lower triangular matrix.

Also, A can be decomposed as
A = UUt ,

(5.29)

315

## in terms of upper triangular form.

For (5.28), the equation Ax = b becomes

Let
then

LLt x = b.

(5.30)

Lt x = z

(5.31)

Lz = b.

(5.32)

The vector z can be obtained from (5.32) by forward substitution and the solution
vector x are determined from the equation (5.31) by back substitution. Also, z and x
can be determined by computing the inverse of L only as
z = L1 b and x = (Lt )1 z = (L1 )t z.

(5.33)

## The inverse of A can be determined as

A1 = (L1 )t L1 .
Procedure to determine L
Since A = LLt , then

l11 0 0 0
l21 l22 0 0 l11 l21 lj1 ln1

0
0

l
A =
j3
n3

## li1 li2 li3 0

0 0 0 lnn
ln1 ln2 ln3 lnn
2
l21 l11
lj1 l11

l11
2 + l2
l21 l11 l21
lj1 l21 + lj2 l22

22

=
li1 l11 l21 li1 + l22 li2 lj1 li1 + + ljj lij

ln1 l11 l21 ln1 + l22 ln2 lj1 ln1 + + ljj lnj

ln1 l11

## ln1 l21 + ln2 l22

.
ln1 li1 + + lni lii

2
2
2
ln1 + ln2 + + lnn

## Equating both sides, we nd the following equations.

l11 = (a11 )1/2


i1
 1/2
2 + l2 + + l2 = a
li1
or
l
=
a

lij
, i = 2, 3, . . . , n
ii
ii
ii
i2
ii
j=1

## li1 = ai1 /l11 , i = 2, 3, . . . , n

li1 lj1 + li2 lj2 + + lij ljj = aij

1/2
j1

1
aij
or, lij = ljj
ljk lik
,
k=1

lij = 0, i < j.

(5.34)
for i = j + 1, j + 2, . . . , n

## 316 Numerical Analysis

Similarly, for the system of equations (5.29), the elements uij of U are given by
unn = (ann )1/2
uin = ain
/unn , i = 1, 2, . . . , 
n1
n

uij = u1jj aij
uik ujk ,
k=j+1

for i = n 2, n 3, . . . , 1; j = i + 1, i + 2, . . . , n 1

1/2
n

2
uik
, i = n 1, n 2, . . . , 1
uii = aii
k=i+1

uij = 0, i > j.
Example 5.9.1 Solve the following system of equations by Cholesky method.
2x1 + x2 x3 = 6
x1 3x2 + 5x3 = 11
x1 + 5x2 + 4x3 = 13.
Solution. The given system of equations is

2 1 1
Ax = b where x = (x1 , x2 , x3 )t , b = (6, 11, 13)t , and A = 1 3 2 .
1 2 4
It is observed
that
A
is
symmetric
and
positive
denite.

l11 0 0
Let L = l21 l22 0 .
l31 l32 l33
2

l11
l11 l21
l11 l31
2 1 1
2 + l2
l21 l31 + l22 l32 = 1 3 2 .
Therefore, LLt = l21 l11 l21
22
2 + l2 + l2
1 2 4
l31 l11 l31 l21 + l32 l22 l31
32
33
Comparing both sides,
we
have

2 = 2 or l
2
l11
11 =
l11 l21 = 1 or l21 = 1/2
l11 l31 = 1 or l31 = 1/ 2
+
1 1/2
2 + l2 = 3 or l
=
l21
22 = (3 2 )
22

5
2

l31 l21 ) = 3/ 10
+
13
2 + l2 + l2 = 4 or l
2 l2 )1/2 =
=
(4

l
l31
33
32
33
31
32
5 .

2
0
0
1.41421
0
0

.
0
Therefore, L = 1/2
5/2  0 = 0.70711 1.58114
0.70711 0.94868 1.61245
1/ 2 3/ 10 13/5
l31 l21 + l32 l22 = 2 or l32 =

1
l22 (2

(5.35)

317

## From the relation Lz = b, we have

1.41421z1 = 6
0.70711z1 + 1.58114z2 = 11
0.70711z1 + 0.94868z2 + 1.61245z3 = 13.
This gives z1 = 4.24265, z2 = 5.05963, z3 = 3.22491.
Now, from the relation Lt x = z, we have

## 1.41421 0.70711 0.70711

4.24265
x1

0
1.58114 0.94868 x2 = 5.05963 .
x3
0
0
1.61245
3.22491
i.e.,

## 1.41421x1 + 0.70711x2 + 0.70711x3 = 4.24265

1.58114x2 + 0.94868x3 = 5.05963
1.61245x3 = 3.22491

## Solution of these equations is x3 = 2.00001, x2 = 1.99993, x1 = 0.99998.

Hence the solution is x1 = 1.0000, x2 = 2.0000, x3 = 2.0000, correct up to four
decimal places.

5.10

## Matrix Partition Method

When a matrix is very large and it is not possible to store the entire matrix into the
primary memory of a computer at a time, then matrix partition method is used to
nd the inverse of a matrix. When a few more variables and consequently a few more
equations are added to the original system then also this method is very useful.
Let the coecient matrix A be partitioned as

..
B
.
C

A=
(5.36)

..
D . E
where B is an l l matrix, C is an l m matrix, D is an m l and E is an m m
matrix; and l, m are positive integers with l + m = n.
Let A1 be partitioned as

..
P
.
Q

(5.37)
A1 =

..
R . S

## 318 Numerical Analysis

where the matrices P, Q, R and S are of the same orders as those of the matrices B, C, D
and E respectively. Then

AA1

..
B . C P

=

..
D . E
R

..
. Q I1

=
..
0
. S

..
. 0

,
..
. I2

(5.38)

where I1 and I2 are identity matrices of order l and m respectively. From (5.38), we
have
BP + CR = I1
BQ + CS = 0
DP + ER = 0
DQ + ES = I2 .
Now, BQ + CS = 0 gives Q = B1 CS i.e., DQ = DB1 CS.
Also, from DQ + ES = I2 , we have (E DB1 C)S = I2 .
Therefore, S = (E DB1 C)1 .
Similarly, the other matrices are
Q = B1 CS
R = (E DB1 C)1 DB1 = SDB1
P = B1 (I1 CR) = B1 B1 CR.
It may be noted that, to nd the inverse of A, it is required to determine the inverses
of two matrices B and (E DB1 C) of order l l and m m respectively.
That is, to compute the inverse of the matrix A of order n n, the inverses of two
lower order (roughly half) matrices are to be determined. If the matrices B, C, D, E
are still large to t in the computer memory, then further partition them.

3 3 4
Example 5.10.1 Find the inverse of the matrix A = 2 1 1 using the matrix
1 3 5
partition method. Hence nd the solution of the system of equations
3x1 + 3x2 + 4x3 = 5
2x1 + x2 + x3 = 7
x1 + 3x2 + 5x3 = 6.

## Solution of System of Linear Equations

Solution. Let the matrix A be partitioned as

3 3 .. 4
..

..

B . C
3 3
2
1
.
1

A=
= , where B = 2 1

.
D .. E
..
1
3 . 5


 
4
C=
, D= 1 3 , E= 5
1

..
P . Q

and A1 =
, where P, Q, R and S are given by
..
R . S
S = (E DB1 C)1 , R = SDB1 , P = B1 B1 CR, Q = B1 CS.
Now,
B1 =

E DB

C =
S =
R =
P =
=
Q =

1
3

1 3
2 3

1
3

1 3
.
2 3

 1 1 3
1
4
5 1 3
= .
2 3
1
3
3
3


 1 1 3


3 1 3
= 5 6
2 3
3


1 1 3
1 1 3
4 
1
1
5 6
B B CR =

2 3
2 3
1
3
3

2
3
.
9 11

1 1 3
4
1
3=

2 3
1
5
3


Therefore,
A1
Hence,

2
3 1
= 9 11 5 .
5
6 3

2
3 1
5
17
x = A1 b = 9 11 5 7 = 62 .
5
6 3
6
35

319

5.11

## If the system of equations is of the form

b1 x1 + c1 x2 = d1
a2 x1 + b2 x2 + c2 x3 = d2

(5.39)

a3 x2 + b3 x3 + c3 x4 = d3

an xn1 + bn xn = dn ,
then the coecient matrix is

b1
a2

0
A=

0
0

c1
b2
a3

0
0

0
c2
b3

0
0

0
0
c3

0
0

0
0
0
0

d1
0
0
0
0

d2

0
0
0
0

and d = .

..

## 0 an1 bn1 cn1

dn
0
0
an bn

(5.40)

It may be noted that the main diagonal and the adjacent coecients on either side of
it consist of only non-zero elements and all other elements are zero. The matrix is called
tri-diagonal matrix and the system of equations is called a tri-diagonal system. These
type of matrices occur frequently in the solution of ordinary and partial dierential
equations by nite dierence method.
A tri-diagonal system can be solved using LU decomposition method.
Let A

= LU where
0
0
1 0 0 0
2 2 0 0
0
0

L =
,
0 0 0 n1 n1 0
0 0 0 0
n n

0
1 1 0 0 0
0 1 2 0 0
0

and U =
.
0 0 0 0 1 n1
0 0 0 0 0
1
Then

1 1
0
0 0
0
1

2 1 2 + 2
2 2
0 0
0

3
2 3 + 3 0 0
0
LU = 0

0
0
0
0 n n n1 + n

## Solution of System of Linear Equations

321

Now, comparing the matrix LU with A and obtain the non-zero elements of L and
U as
1 = b1 ,

i i = ci ,

i = ai , i = 2, . . . , n
i = bi i1 i = bi ai

i = 1, 2, . . . , n 1

or, i = ci /i ,
ci1
,
i1

i = 2, 3, . . . , n.

## Thus the elements of L and U are given by the following relations.

1 = b1 ,

ai ci1
, i = 2, 3, . . . , n
i1
i = ai , i = 2, 3, . . . , n
i = bi

(5.41)
(5.42)

i = ci /i , i = 1, 2, . . . , n 1.
The solution of the equation (5.39) i.e., Ax = d where d = (d1 , d2 , . . . , dn )t can be
obtained by solving Lz = d using forward substitution and then solving Ux = z using
back substitution. The solution of Lz = d is given by
z1 =

d1
di ai zi1
, zi =
, i = 2, 3, . . . , n.
b1
i

(5.43)

## The solution of the equation Ux = z is

xn = zn , xi = zi i xi+1 = zi

ci
xi+1 ,
i

i = n 1, n 2, . . . , 1.

## Example 5.11.1 Solve the following tri-diagonal system of equation.

= 3,
x1 + 2x2 + x3 = 6,
3x2 + 2x3 = 12.
x1 + x2
Solution. Here b1 = c1 = 1, a2 = 1, b2 = 2, c2 = 1, a3 = 3, b3 = 2,
d1 = 3, d2 = 6, d3 = 12.
Therefore,
1 = b1 = 1

c1
= 2 (1).1 = 3
1
1
c2
3 = b3 a3 = 2 3. = 1
2
3
d1
d 2 a2 z 1
= 3,
z2 =
= 3,
z1 =
b1
2
c2
x2 = z2 x3 = 2,
x3 = z3 = 3,
2
2 = b2 a2

d3 a3 z2
=3
3
c1
x1 = z1 x2 = 1.
1
z3 =

(5.44)

## 322 Numerical Analysis

It may be noted that the equation (5.43) and also (5.44) are valid only if i = 0
for all i = 1, 2, . . . , n. If any one of i becomes zero at any stage then the method is
not applicable. Actually, this method is based on LU decomposition technique and LU
decomposition method is applicable and unique if principal minors of coecient matrix
of all orders are non-zero. But, if a minor becomes zero then a modication on (5.43)
and (5.44) gives the solution of the tri-diagonal system.
Suppose k = 0 and i = 0, i = 1, 2, . . . , k 1. Then let k = s, a symbolic value of k .
Then the remaining i , i = k + 1, . . . , n are calculated using the equation (5.41). These
s are used to calculate zi and xi using the formulae (5.43) and (5.44). The values of
xi s are obtained in terms of s. Lastly, the nal solution is obtained by substituting
s = 0. The following example illustrates this case.
Example 5.11.2 Solve the following system of equations.
x1 + x2
= 3,
x1 + x2 3x3 = 3,
2x2 + 3x3 = 4.
Solution. Here b1 = c1 = 1, a2 = b2 = 1, c2 = 3, a3 = 2, b3 = 3,
d1 = 3, d2 = 3, d3 = 4.
Therefore,
1 = b1 = 1
2 = b2 a2

c1
=11=0
1

c2
6
3
=3
=3+2
2
s
s
d1
d 2 a2 z 1
6
= 3,
z2 =
= ,
b1
2
s
4s 12
z3 =
,
3s 6
c2
6
,
z2 x3 =
2
3s 6
c1
9s 12
.
z1 x2 =
1
3s 6

3 = b3 a3
z1 =
x3 =
x2 =
x1 =

z3 =

d 3 a3 z 2
4s 12
=
3
3s 6

## Substituting s = 0 to nd the solution. The required solution is

x1 = 2, x2 = 1, x3 = 2.

323

## Algorithm 5.7 (Solution of a tri-diagonal system). This algorithm solves a

tri-diagonal system of linear equations. Assume that the principal minors are nonzero.
Algorithm Tridiagonal
//Let the tri-diagonal system is of the form Ax = d where A and d are given by
(5.40).//
Step 1. Read the matrix A, i.e., the arrays ai , bi , ci , i = 2, 3, . . . , n1
and b1 , c1 , an , bn .
ai ci1
Step 2. Compute 1 = b1 , i = bi
, i = 2, 3, . . . , n.
i1
d1
di ai zi1
, i = 2, 3, . . . , n.
Step 3. Compute z1 = , zi =
b1
i
ci
Step 4. Compute xn = zn , xi = zi xi+1 , i = n 1, n 2, . . . , 1.
i
Step 5. Print xi , i = 1, 2, . . . , n and Stop.
end Tridiagonal
Program 5.7
.
/* Program TriDiagonal
Program to solve a tri-diagonal system of equations.
The coefficient matrix are taken as a[i],b[i],c[i],
i=2, 3, ..., n-1, b,c,a[n],b[n]. The right
hand vector is d[i], i=1, 2, ..., n.*/
#include<stdio.h>
#include<stdlib.h>
float x; /* x[i]is the solution of the tri-diagonal system */
void main()
{
float a,b,c,d;
int i,n; float y;
float TriDiag(float [],float [],float [],float [],int);
printf("Enter the size of the coefficient matrix ");
scanf("%d",&n);
printf("Enter first row (only non-zero elements) ");
scanf("%f %f",&b,&c);
printf("Enter rows 2 to n-1 ");
for(i=2;i<=n-1;i++) scanf("%f %f %f",&a[i],&b[i],&c[i]);
printf("Enter last row ");
scanf("%f %f",&a[n],&b[n]);

## printf("Enter the right hand vector ");

for(i=1;i<=n;i++) scanf("%f",&d[i]);
y=TriDiag(a,b,c,d,n);/* call of TriDiag to a dummy variable */
printf("The solution is \n");
for(i=1;i<=n;i++) printf("%f ",x[i]);
} /* end of main */
float TriDiag(float a,float b,float c,float d,int n)
{
/* output x[i], i=1, 2,..., n, is a global variable.*/
int i; float gamma,z;
gamma=b;
for(i=2;i<=n;i++)
{
if(gamma[i-1]==0.0)
{
printf("A minor is zero: Method fails ");
exit(0);
}
gamma[i]=b[i]-a[i]*c[i-1]/gamma[i-1];
}
z=d/gamma;
for(i=2;i<=n;i++)
z[i]=(d[i]-a[i]*z[i-1])/gamma[i];
x[n]=z[n];
for(i=n-1;i>=1;i--)
x[i]=z[i]-c[i]*x[i+1]/gamma[i];
/* for(i=1;i<=n;i++) printf("%f ",x[i]); */
return(x);
} /*end of TriDiag */
A sample of input/output:
Enter the size of the coefficient matrix 4
Enter first row (only non-zero elements) 1 2
Enter rows 2 to n-1
3 2 1
2 0 -1
Enter last row
1 2

325

## Enter the right hand vector

3 2 1 1
The solution is
0.500000 1.250000 -2.000000 1.500000

5.12

## For n 3, a general tri-diagonal matrix T = [tij ]nn is of the form

0
b1 c1 0
a2 b2 c2
0

0 a3 b3
0

T=

0 0 0 an1 bn1 cn1
0 0 0 0
an bn
for any tij = 0 for |i j| 2.
The entire matrix can be stored using only three vectors c = (c1 , c2 , . . . , cn1 ), a =
(a2 , a3 , . . . , an ), and b = (b1 , b2 , . . . , bn ). We dene a vector d = (d1 , d2 , . . . , dn ) as
,
if i = 1
b1
di = b ai c
(5.45)
i
i1 if i = 2, 3, . . . , n.
di1
If di = 0 for any i n, then, set di = x (x is just a symbolic name) and continue to
compute di+1 , di+2 , . . . , dn in terms of x by using (5.45).
n

di (in general, this is a polynomial in x) evaluated at x = 0 is
The product P =
i=1

the value of |T|. If P is free from x then the product P directly gives the value of |T|.
Example 5.12.1 Find the values of the determinants of the following tri-diagonal
matrices.

1 1 0
1 1 0
A = 1 1 2 ,
B = 1 2 1 .
0 3 4
0 1 2
Solution. For the matrix A,

a3
4x 6
.
c2 =
d2
x
Therefore, P = d1 d2 d2 = 4x 6, gives |A| = 6.

d1 = 1, d2 = 0, so, set d2 = x, d3 = b3

1
M.E.A.El-Mikkawy, A fast algorithm for evaluating nth order tri-diagonal determinants, J. Computational and Applied Mathematics, 166 (2004) 581-584.

## 326 Numerical Analysis

For the matrix B,
a2
a3
d1 = 1, d2 = b2 c1 = 1, d3 = b3 c2 = 1.
d1
d2
Therefore, P = d1 d2 d2 = 1.1.1 = 1, gives |B| = 1.

5.13

## Vector and Matrix Norms

The norm of a vector is the size or length of that vector. The norm of a vector x is
denoted by x. This is a real number and satises the following conditions
(i) x 0 and x = 0 i x = 0

(5.46)

(5.47)

(5.48)

## Let x = (x1 , x2 , . . . , xn )t be any vector. The commonly used norms are

(i) x1 =

n


|xi |

(5.49)

i=1

. n
.
|xi |2 (Euclidean norm)
(ii) x2 = /

(5.50)

i=1

## (iii) x = max |xi | (maximum norm or uniform norm).

i

(5.51)

Let A and B be two matrices such that A + B and AB are dened. The norm of
a matrix A = [aij ] is denoted by A, which satises the following conditions
(i) A 0 and A = 0 i A = 0

(5.52)

(5.53)

(5.54)

(5.55)

## From (5.55), it can be veried that

Ak  Ak ,
for any positive integer k.

(5.56)

## Like the vector norms, the matrix norms may be dened as


(i) A1 = max
|aij | (the column norm)
j

(ii) A2 =

327

(5.57)

0 
i

(5.58)

(5.59)

i


j

## The Euclidean norm is also known as Erhard-Schmidt norm or Schur norm or

the Frobenius norm.
The concept of matrix norm is used to study the stability of a system of equations.
It is also used to study the convergence of iterative methods to solve the linear system
of equations.
Example
5.13.1
Find the matrix norms A1 , A2 and A for the matrix

2 3 4
A = 0 1 5 .
3 2 6
Solution.
A1 = max{2 + 0 + 3, 3 1 + 2, 4 + 5 + 6} = 15


## A2 = 22 + 32 + 42 + 02 + (1)2 + 52 + 32 + 22 + 62 = 104 and

A = max{2 + 3 + 4, 0 1 + 5, 3 + 2 + 6} = 11.

5.14

## Before introduction of ill-conditioned system, let us consider the following system of

equations.
x + 3y = 4
1
x + y = 1.33.
3

(5.60)

Note that this system of equations has no solution. But, if we take the approximate
value of 13 as 0.3 then (5.60) becomes
x + 3y = 4
0.3x + y = 1.33.

(5.61)

## 328 Numerical Analysis

The solution of (5.61) is x = 0.1, y = 1.3.
If the approximation of 13 is taken as 0.33 then the solution of the system of equations
x + 3y = 4
0.33x + y = 1.33
is x = 1, y = 1.
The approximations 0.333 and 0.3333 of

1
3

(5.62)

## give the following systems

x + 3y = 4
0.333x + y = 1.33.

(5.63)

## whose solution is x = 10, y = 2 and

x + 3y = 4
0.3333x + y = 1.33.

(5.64)

## with solution x = 100, y = 32.

The systems (5.60)-(5.64) and their solutions indicate a dangerous situation. It may
be noted that the dierent approximations of 13 give high variations in their solutions.
What is conclusion about the above systems? We may conclude that the systems are
unstable. That is, a small change in the coecients of the system produces large change
in the solution. These systems are called ill-conditioned or ill-posed system. On the
other hand, if the change in the solution is small for small changes in the coecients,
then the system is called well-conditioned or well-posed system.
Let the system of equations be
Ax = b.

(5.65)

## Let A and b be matrices obtained from A and b by introducing small changes in

A and b and let y be the solution of the new system. That is,
A y = b .

(5.66)

The system (5.65) is called ill-conditioned when the changes in y are too large compared to those in x. Otherwise, the system is called well-conditioned. If a system is
ill-conditioned then the corresponding coecient matrix is called an ill-conditioned
matrix.

1 3
The system (5.62) is ill-conditioned and the corresponding coecient matrix
0.33 1
is an ill-conditioned matrix.
Generally, ill-condition occurs when |A| is small. To measure the ill-condition of a
matrix, dierent methods are available. One of the useful method is introduced here.

## Solution of System of Linear Equations

329

The quantity Cond(A), called the condition number of the matrix, dened by
Cond(A) = A A1 

(5.67)

where A is any matrix norm, gives a measure of the condition of the matrix
A. The large value of Cond(A) indicates the ill-condition of a matrix or the associated
system of equations.

1 3
4 3
Let A =
and B =
be two matrices.
0.33 1
3 5

5 3
100 300
1
1
1
.
Then A =
and B = 11
3 4
33 100

## The Euclidean norms, A2 = 1 + 9 + 0.10890 + 1 = 3.3330 and A1 2 = 333.3001.

Therefore, Cond(A) = A2 A1 2 = 1110.88945, a very large number. Hence A
is ill-conditioned.
Where as B2 = 7.68115 and B1 2 = 0.69829
Then Cond(B) = 5.36364, a relatively small quantity.
Therefore, B is well-conditioned matrix.
Another indicator of ill-conditioning matrix is presented below.

1/2
n
Let A = [aij ] be the matrix and ri =
a2ij
. The quantity
j=1

(A) =

|A|
r1 r2 r n

(5.68)

## measures the smallness of the determinant of A. If is very small compared to 1, then

the matrix A is ill-conditioned,

otherwise A is well-conditioned.

1 3
For the matrix A =
, r1 = 10, r2 = 1.05304, |A| = 0.01,
0.33 1

0.01
4 3
= 0.003 and for B =
, r1 = 5, r2 = 34,
(A) =
3 5
10 1.05304
11
= 0.37730.
|B| = 11, (B) =
5 34
Hence A is ill-conditioned while B is well-conditioned.
5.14.1

## Method to solve ill-conditioned system

Some methods are available to solve an ill-conditioned system of linear equations. One
straight forward technique is to carry out the calculations with more number of significant digits. But, the calculations with more signicant digits is time-consuming. One
suggested method is to improve upon the accuracy of the approximate solution by an
iterative method. This iterative method is discussed below.

## 330 Numerical Analysis

Let the system of equations be
n


aij xj = bi , i = 1, 2, . . . , n.

(5.69)

j=1

12 , . . . , x
1n be an approximate solution of (5.69). Since this is an approximate
Let x
11 , x
n

aij x
1j is not necessarily equal to bi . Let bi = 1bi for this approximate solution.
solution,
j=1

## Then, for this solution, (5.69) becomes

n


aij x
1j = 1bi , i = 1, 2, . . . , n.

(5.70)

j=1

## Subtracting (5.70) from (5.69), we obtain

n


i.e.,

j=1
n


aij (xj x
1j ) = (bi 1bi )
aij i = di

(5.71)

j=1

1i , di = bi 1bi , i = 1, 2, . . . , n.
where i = xi x
Now, the solution for i s is obtained by solving the system (5.71). Hence the new
solution is given by xi = i + x
1i and these values are better approximations to xi s. This
technique can be repeated again to improve the accuracy.

5.15

## Generalized Inverse (g-inverse)

The conventional matrix inverse (discussed in Section 5.3) is widely used in many areas
of science and engineering. It is also well known that conventional inverses can be
determined only for square non-singular matrices. But, in many areas of science and
engineering such as statistics, data analysis etc. some kind of weak inverses of singular
square and rectangular matrices are very much essential. The inverses of such types of
matrices are known as generalized inverse or g-inverse. A number of works have
been done during last three decades on g-inverse. The generalized inverse of an m n
matrix A is a matrix X of size n m. But, dierent types of generalized inverses are
dened by various authors. The following matrix equations are used to classify the
dierent types of generalized inverses for the matrix A:
(i)
(ii)
(iii)
(iv)

AXA = A
XAX = X
AX = (AX)
XA = (XA) , ( denotes the conjugate transpose).

(5.72)

331

## The matrix X is called

(a) a generalized inverse of A, denoted by A , if (i) holds;
(b) a reexive generalized inverse of A, denoted by A
r , if (i) and (ii) hold ;
(c) a minimum norm inverse of A, denoted by A
,
if
(i) and (iv) hold;
m

## (d) a least-square inverse of A, denoted by Al , if (i) and (iii) hold;

(e) the Moore-Penrose inverse of A, denoted by A+ , if (i), (ii), (iii) and (iv) hold.
Only the Moore-Penrose inverse A+ is unique and other all inverses are not unique.
One interesting result is that, when A is non-singular then all these inverses reduce to
A1 . Due to the uniqueness of A+ , this generalized inverse is widely used.
Some important properties are presented below:
(i) (A+ )+ = A;
+ t

(5.73)

t +

(ii) (A ) = (A ) ;
(iii) A

(5.74)

= A , if A is non-singular

(5.75)

+

(5.76)

## Let A be a matrix of order m n.

If rank of A is 0, then A+ is a null matrix of order n m;
1
At ;
(vi) If rank of A is 1, then A+ =
trace(AAt )
(vii) If rank of A is n, then A+ = (At A)1 At ;
(v)

t 1

(5.77)
(5.78)
(5.79)
(5.80)

## The g-inverse A+ of the matrix A is used to solve the system of equations Ax =

b, (b = 0) where A is an m n matrix, x and b are respectively n 1 and m 1
vectors. The solution of Ax = b is given by
x = A+ b

(5.81)

5.15.1

## Greville4 presented a recursive algorithm to nd the Moore-Penrose inverse of a matrix.

The Euclidean norm x = x x is minimum for any choice of arbitrary inverse.
3
Least-squares solution minimizes Ax b for an inconsistent system.
4
For the proof of the algorithm see Greville, T.N.E, The pseudo-inverse of a rectangular or singular
matrix and its application to the solution of system of linear equations, SIAM Review, 1 (1959) 38-43.
2

Let

## a11 a12 a1k a1n

a21 a22 a2k a2n

(5.82)
A=
= (1 2 . . . k . . . n )
am1 am2 amk amn

a1k
a2k

## where k = . , the kth column of the matrix A.

.
.
amk
Also, let Ak be the matrix formed by the rst k columns, i.e., Ak = (1 2 . . . k ).
Then
Ak = (Ak1 k ).

## The proposed algorithm is recursive and recursion is started with

A+
1 = 0 if 1 = 0 (null column) else

t
1 t
A+
1 .
1 = (1 1 )

(5.83)

## Let the column vectors be

k = A+
k1 k

(5.84)

and k = k Ak1 k .

(5.85)

If k = 0, then compute
else

k = k+ = (kt k )1 kt ;

(5.86)

k = (1 +

(5.87)

k is given by
A+
k =

kt k )1 kt A+
k1 .

A+
k1 k k .
k

(5.88)

## The process is repeated for k = 1, 2, . . . , n.

Example 5.15.1 Obtain

## the g-inverse (Moore-Penrose

2 1 0 1
1 0 1 0
0 1 1 2
and use it to solve the following system of equations
+ x4 = 4
2x1 + x2
x3
=0
x1
x2 + x3 + 2x4 = 4.

inverse)

of

## Solution of System of Linear Equations

2
1
0
1
2

Solution. Here 1 = 1 , 2 = 0 , 3 = 1 , 4 = 0 , A1 = 1 .
0
1
1
2
0

1




 

 2
t )1 t = 2 1 0 1
2 1 0 = 15 2 1 0 = 25 15 0
=
(
A+
1 1
1
1
0

2 1  1
 
0 = 2
2 = A+

=
1 2
5 5 0
5
1
1

1
2  
5
2 = 2 A1 2 = 0 1 25 = 25 = 0 (the null column vector).
1
0
1
1 1

5

1 2 

Hence 2 = 2+ = (2t 2 )1 2t = 15 25 1 25
5 5 1
1
 


= 56 15 25 1 = 16 13 56

  1

2 1
2 2 = 25 16 13 56 = 15
15
3

1 1 1

A1 2 2
+
= 13 31 35
A2 =
2

6
3
6

1 1 1

3
+
3
3
3
Now, 3 = A2 3 = 1 1 5 1 =
7
6 3
6
6
1
1

0
2 1
6
23
= 13 = 0.
3 = 3 A2 3 = 1 1 0
7
6
1
0 1
16




Hence 3 = 3+ = (3t 3 )1 3t = 6 16 13 16 = 1 2 1
2

2 4 2


3 
3 3 3
1
2
1
3 3 =
=
7
7
7
7
6
6 3 6

+
1 1 1
A2 3 3
= 1 2 2
A+
3 =
3
1 2 1

1 1 1
1
1
1 2 2 0 = 3 ,
Now, 4 = A+
3 4 =
1 2 1
2
1

0
1
1
2 1 0
1
1

0
0
1
0
1
3
0
= 0.

4 = 4 A3 4 =
0
2
2
0 1 1
1
2
(the null column vector)

333

## 334 Numerical Analysis

1 1 1

1
1 2 2 = 5
4 = (1 + 4t 4 )1 4t A+
3 = 12 1 3 1
12
1 2 1

## 5/12 3/4 2/3

1 

5 3 2

2
3
4 4 =
12 4 3 = 5/4 9/4
5/12 3/4 2/3
1

7/12
1/4
1/3

+
1/4 1/4 0
A3 4 4
+

=
=
A+
4
7/12 5/4 1/3 = A
4
5/12 3/4 2/3
+
The given system
solution isx =
is Ax = b and its
A b.

## 7/12 1/4 1/3

1
1/4 1/4 0 4
1

Therefore, x =
7/12 5/4 1/3 0 = 1
4
5/12 3/4 2/3
1
Therefore the required solution is
x1 = 1, x2 = 1, x3 = 1, x4 = 1.
So that

3 2
4 3

Note 5.15.1 It may be veried that A+ satises all the conditions (5.72). Again,
1
as |A3 | = 1 i.e., A3 is non-singular. In addition to this, for this A,
A+
3 = A3
+
AA = I4 , but A+ A = I4 , the unit matrix of order 4.
The g-inverse A+ of A can also be computed using the formula (5.80).

5.16

## Let the system of linear equations be

Ax = b

(5.89)

where A, x and b are of order mn, n1 and m1 respectively. Here, we assume that
(5.89) is inconsistent. Since (5.89) is inconsistent, the system has no solution. Again,
since there may be more than one xl (least-squares solutions) for which Ax b is
minimum, there exist one such xl (say xm ) whose norm is minimum. That is, xm is
called minimum norm least squares solution if
xm  xl 

(5.90)

Axl b Ax b

for all x.

(5.91)

## Solution of System of Linear Equations

335

The minimum norm least squares solution can be determined using the relation
x = A+ b.

(5.92)

## Since A+ is unique, the minimum norm least squares solution is unique.

The least squares solution can also be determined by the following way.
The vector Ax b in terms of the elements of A, x and b is

## a11 x1 + a12 x2 + a1n xn b1

a21 x1 + a22 x2 + a2n xn b2
.

am1 x1 + am2 x2 + amn xn bm

## Let square of Ax b be denoted by S. Then

S = (a11 x1 + a12 x2 + a1n xn b1 )2
+(a21 x1 + a22 x2 + a2n xn b2 )2 +
+(am1 x1 + am2 x2 + amn xn bn )2
n
m 

=
(aij xj bi )2 .

(5.93)

i=1 j=1

## Also, S is called the sum of square of residues. To solve (5.89), nd x = (x1 , x2 , . . . , xn )t

from (5.93) in such a way that S is minimum. The conditions for S to be minimum are
S
S
S
= 0,
= 0, ,
=0
x1
x2
xn

(5.94)

## The system (5.94) is non-homogeneous and consists of n equations with n unknowns

x1 , x2 , . . . , xn . This system can be solved by any method. Let x1 = x1 , x2 = x2 , . . . , xn =
xn be a solution of (5.94). Therefore, the least-squares solution of (5.89) is
x = (x1 , x2 , . . . , xn )t

(5.95)

## and the sum of square of residues is given by

S =

n
m 


(aij xj bi )2 .

(5.96)

i=1 j=1

This method is not suitable for a large system of equations, while the method stated
in equation (5.92) is applicable for a large system also.

## 336 Numerical Analysis

Example 5.16.1 Find g-inverse of the singular matrix A =

3 6
2 4

and hence nd

## a least squares solution of the inconsistent system

3x + 6y = 9
2x + 4y = 5.

3
6
3
, 2 =
, A1 =
.
Solution. Let 1 =
2
4
2




3 2
t
1 t = 1 3 2 =
A+
1
1 = (1 1 )
13

13 13 ,
 3 2  6
2 = A+
= 2,
1 2 = 13 13

4

6
3
0
2 = 2 A1 2 =

.2 =
= 0 (a null vector),
4
2
0


 6 4 
3 2
= 15 .2. 13
=
2 = (1 + 2t 2 )1 2t A+
1
13
65 65


8
2 2 = 12
65 65

3 2

+
A1 2 2
+
65
= 65
= A+ , which is the g-inverse of A.
Therefore, A2 =
6 4
2
65 65
Second Part: The given
equations

can
be written

as
3 6
x
9
Ax = b, where A =
, x=
, A=
.
2 4
y
5
Then the least squares solution

is
given by

3
2
1
9
1
x = A+ b, i.e., x = 65
= 37
65 2 .
6 4
5
74
37
Hence the least squares solution is x = , y = .
65
65
Example 5.16.2 Find the least squares solution of the following equations x + y =
3.0, 2x y = 0.03, x + 3y = 7.03, and 3x + y = 4.97. Also, estimate the residue.
Solution. Let x, y be least squares solution of the given system. Then the square of
residues S is
S = (x + y 3.0)2 + (2x y 0.03)2 + (x + 3y 7.03)2 + (3x + y 4.97)2 .
We choose x and y in such a way that S is minimum. Therefore
S
S
= 0 and
= 0.
x
y
That is,
2(x + y 3.0) + 4(2x y 0.03) + 2(x + 3y 7.03) + 6(3x + y 4.97) = 0
and 2(x + y 3.0) 2(2x y 0.03) + 6(x + 3y 7.03) + 2(3x + y 4.97) = 0.

337

## These equations reduce to 3x + y 5 = 0 and 5x + 12y 29.03 = 0.

Solution of these equations is x = 30.97/31 = 0.9990322 and y = 62.09/31 =
2.0029032, which is the required solution of the given equations.
The sum of the square of residues is S = (3.0019354 3.0)2 + (0.0048388 0.03)2 +
(7.0077418 7.03)2 + (4.9999998 4.97)2 = 0.0026129.

Iteration Methods
If the system of equations has a large number of variables, then the direct methods
are not much suitable. In this case, the approximate numerical methods are used to
determine the variables of the system.
The approximate methods for solving system of linear equations make it possible to
obtain the values of the roots of the system with the specied accuracy as the limit of
the sequence of some vectors. The process of constructing such a sequence is known as
the iterative process.
The eciency of the application of approximate methods depends on the choice of
the initial vector and the rate of convergence of the process.
The following two approximate methods are widely used to solve a system of linear
equations:
(i) method of iteration (Jacobis iteration method), and
(ii) Gauss-Seidals iteration method.
Before presenting the iteration methods, some terms are introduced to analyse the
methods.
(k)
Let xi , i = 1, 2, . . . , n be the kth (k = 1, 2, . . .) iterated value of the variable xi and
(k) (k)
(k)
x(k) = (x1 , x2 , . . . , xn )t be the solution vector obtained at the kth iteration.
(k)
The sequence {x }, k = 1, 2, . . . is said to converge to a vector x = (x1 , x2 , . . . , xn )t
if for each i (= 1, 2, . . . , n)
(k)

xi

xi as k .

(5.97)

## Let = (1 , 2 , . . . , n )t be the exact solution of the system of linear equations. Then

(k)
the error i in the ith variable xi committed in the kth iteration is given by
(k)

(k)

= i xi .

(5.98)

## The error vector (k) at the kth iteration is then given by

(k)

(k)

t
(k) = (1 , 2 , . . . , (k)
n ) .

(5.99)

## The error dierence e(k) at two consecutive iterations is given by

e(k) = x(k+1) x(k) = (k) (k+1) ,

(5.100)

## 338 Numerical Analysis

(k)

(k+1)

(k)

where ei = xi
xi .
An iteration method is said to be of order p 1 if there exists a positive constant A
such that for all k
(k+1)  A(k) p .

5.17

(5.101)

## Let us consider a system of n linear equations containing n variables:

a11 x1 + a12 x2 + + a1n xn = b1
a21 x1 + a22 x2 + + a2n xn = b2

(5.102)

an1 x1 + an2 x2 + + ann xn = bn .
Also, we assume that the quantities aii are pivot elements.
The above equations can be written as
1
(b1 a12 x2 a13 x3 a1n xn )
a11
1
(b2 a21 x1 a23 x3 a2n xn )
x2 =
a22

1
(bn an1 x1 an2 x2 an n1 xn1 ).
xn =
ann
x1 =

(0)

(0)

(5.103)

(0)

## Let x1 , x2 , . . . , xn be the initial guess to the variables x1 , x2 , . . . , xn respectively

(initial guess may be taken as zeros). Substituting these values in the right hand side
of (5.103), which yields the rst approximation as follows.
1
(0)
(0)
(b1 a12 x2 a13 x3 a1n x(0)
n )
a11
1
(1)
(0)
(0)
(b2 a21 x1 a23 x3 a2n x(0)
x2 =
n )
a22

1
(0)
(0)
(0)
=
(bn an1 x1 an2 x2 an n1 xn1 ).
x(1)
n
ann
(1)

x1

(1)

(1)

(1)

(5.104)

Again, substituting x1 , x2 , . . . , xn in the right hand side of (5.103) and obtain the
(2) (2)
(2)
second approximation x1 , x2 , . . . , xn .

## Solution of System of Linear Equations

(k)

(k)

339

(k)

In general, if x1 , x2 , . . . , xn be the kth approximate roots then the next approximate roots are given by
1
(k)
(k)
(b1 a12 x2 a13 x3 a1n x(k)
n )
a11
1
(k+1)
(k)
(k)
x2
=
(b2 a21 x1 a23 x3 a2n x(k)
n )
a22

1
(k)
(k)
(k)
(k+1)
=
(bn an1 x1 an2 x2 an n1 xn1 ).
xn
ann
k = 0, 1, 2, . . . .
(k+1)

x1

(5.105)

The iteration process is continued until all the roots converge to the required number
of signicant gures. This iteration method is called Jacobis iteration or simply the
method of iteration.
The Jacobis iteration method surely converges if the coecient matrix is diagonally
dominant.
5.17.1

## The Gauss-Jacobis iteration scheme (5.105) can also be written as



n

1
(k+1)
(k)
xi
bi
=
aij xj , i = 1, 2, . . . , n.
aii
j=1

(5.106)

j=i



n

1
bi
aij j .
i =
aii
j=1

(5.107)

j=i

(k+1)

i xi

n

1  
(k)
aij j xj
aii j=1
j=i

or

(k+1)

n
1 
(k)
=
aij j .
aii j=1
j=i

That is,
(k+1)

i

1 
1 
(k)
|aij | j 
|aij | (k) .
|aii | j=1
|aii | j=1
n

j=i

j=i

Let

A = max
i

n
 1 

|aij | .
|aii | j=1
j=i

## Then the above relation becomes

(k+1)  A(k) .

(5.108)

This relation shows that the rate of convergence of Gauss-Jacobis method is linear.
Again,
(k+1)  A(k)  A2 (k1)  Ak+1 (0) .
That is,
(k)  Ak (0) .

(5.109)

## If A < 1 then Ak 0 as k and consequently (k)  0 as k , i.e., the

iteration converges.
Hence the sucient condition for convergent of Gauss-Jacobis method is A < 1 i.e.,
n

|aij | < |aii | for all i, i.e., the coecient matrix is diagonally dominant.
j=1
j=i

## The relation (5.108) can be written as

(k+1)  A(k)  = A e(k) + (k+1) 
Ae  + A
A
e(k) .
(k+1) 
A1
(k)

or,

(k+1)

[by (5.100)]


(5.110)

This relation gives the absolute error at the (k + 1)th iteration in terms of the error
dierence at kth and (k + 1)th iterations.
Example 5.17.1 Solve the following system of linear equations by Gauss-Jacobis
method correct up to four decimal places and calculate the upper bound of absolute
errors.
27x + 6y z = 54
6x + 15y + 2z = 72
x + y + 54z = 110.
Solution. Obviously, the system is diagonally dominant as
|6| + | 1| < |27|,
|6| + |2| < |15|,
|1| + |1| < |54|.

341

## The Gauss-Jacobis iteration scheme is


1
54 6y (k) + z (k)
x(k+1) =
27

1
y (k+1) =
27 6x(k) 2z (k)
15

1
(k+1)
z
=
110 x(k) y (k) .
54
Let the initial solution be (0, 0, 0). The next iterations are shown in the following
table.
k
0
1
2
3
4
5
6
7
8
9
10
11

x
0
2.00000
1.00878
1.24225
1.15183
1.17327
1.16500
1.16697
1.16614
1.16640
1.16632
1.16635

y
0
4.80000
3.72839
4.14167
4.04319
4.08096
4.07191
4.07537
4.07454
4.07488
4.07477
4.07481

z
0
2.03704
1.91111
1.94931
1.93733
1.94083
1.93974
1.94006
1.93996
1.93999
1.93998
1.93998

## The solution correct up to four decimal places is

x = 1.1664, y = 4.0748, z = 1.9400.
Here



n
7 8 2
1
8
, ,
|aij | = max
A = max
= .
i
aii j=1
27 15 54
15
j=i

e(0)

= (3

105 , 4

105 , 0).

## Therefore, the upper bound of absolute error is

(0) 

A
e(0)  = 5.71 105 .
1A

## Algorithm 5.8 (Gauss-Jacobis). This algorithm nds the solution of a system

of linear equations by Gauss-Jacobis iteration method. The method will terminate
(k+1)
(k)
xi | < , where is the supplied error tolerance, for all i.
when |xi

## 342 Numerical Analysis

Algorithm Gauss Jacobi
Step 1. Read the coecients aij , i, j = 1, 2, . . . , n and the right hand vector
bi , i = 1, 2, . . . , n of the system of equations and error tolerance .
Step 2. Rearrange the given equations, if possible, such that the system becomes
diagonally dominant.
Step 3. Rewrite the ith equation as


n

1
bi
for i = 1, 2, . . . , n.
aij xj ,
xi =
aii
j=1
j=i

## Step 4. Set the initial solution as

xi = 0, i = 1, 2, 3, . . . , n.
Step 5. Calculate the new values xni of xi as


n

1
xni =
bi
for i = 1, 2, . . . , n.
aij xj ,
aii
j=1
j=i

Step 6. If |xi xni | < ( is an error tolerance) for all i, then goto Step 7 else
set xi = xni for all i and goto Step 5.
Step 7. Print xni , i = 1, 2, . . . , n as solution.
end Gauss Jacobi
Program 5.8
.
/*Program Gauss_Jacobi
Solution of a system of linear equations by Gauss-Jacobis iteration
method. Testing of diagonal dominance is also incorporated.*/
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
float a,b,x,xn,epp=0.00001,sum;
int i,j,n,flag;
printf("Enter number of variables ");
scanf("%d",&n);
printf("\nEnter the coefficients rowwise ");
for(i=1;i<=n;i++)
for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("\nEnter right hand vector ");
for(i=1;i<=n;i++)
scanf("%f",&b[i]);
for(i=1;i<=n;i++) x[i]=0; /* initialize */

## /* checking for row dominance */

flag=0;
for(i=1;i<=n;i++)
{
sum=0;
for(j=1;j<=n;j++)
if(i!=j) sum+=fabs(a[i][j]);
if(sum>fabs(a[i][i])) flag=1;
}
/* checking for column dominance */
if(flag==1)
{
flag=0;
for(j=1;j<=n;j++)
{
sum=0;
for(i=1;i<=n;i++)
if(i!=j) sum+=fabs(a[i][j]);
if(sum>fabs(a[j][j])) flag=1;
}
}
if(flag==1)
{
printf("The coefficient matrix is not diagonally dominant\n");
printf("The Gauss-Jacobi method does not converge surely");
exit(0);
}
for(i=1;i<=n;i++) printf(" x[%d] ",i);printf("\n");
do
{
for(i=1;i<=n;i++)
{
sum=b[i];
for(j=1;j<=n;j++)
if(j!=i) sum-=a[i][j]*x[j];
xn[i]=sum/a[i][i];
}
for(i=1;i<=n;i++) printf("%8.5f ",xn[i]);printf("\n");

343

## 344 Numerical Analysis

flag=0; /* indicates |x[i]-xn[i]|<epp for all i */
for(i=1;i<=n;i++) if(fabs(x[i]-xn[i])>epp) flag=1;
if(flag==1) for(i=1;i<=n;i++) x[i]=xn[i]; /* reset x[i] */
}while(flag==1);
printf("Solution is \n");
for(i=1;i<=n;i++) printf("%8.5f ",xn[i]);
} /* main */
A sample of input/output:
Enter number of variables 3
Enter the coefficients rowwise
9 2 4
1 10 4
2 -4 10
Enter right hand vector
20 6 -15
x
x
x
2.22222 0.60000 -1.50000
2.75556 0.97778 -1.70444
2.76247 1.00622 -1.66000
2.73640 0.98775 -1.65000
2.73606 0.98636 -1.65218
2.73733 0.98727 -1.65267
2.73735 0.98733 -1.65256
2.73729 0.98729 -1.65254
2.73729 0.98729 -1.65254
Solution is
2.73729 0.98729 -1.65254

5.18

## A simple modication of Jacobis iteration sometimes give faster convergence. The

modied method is known as Gauss-Seidals iteration method.
Let us consider a system of n linear equations with n variables.
a11 x1 + a12 x2 + + a1n xn = b1
a21 x1 + a22 x2 + + a2n xn = b2

an1 x1 + an2 x2 + + ann xn = bn .

(5.111)

345

## Assume that the diagonal coecients aii , i = 1, 2, . . . , n are diagonally dominant. If

this is not the case then the above system of equations are re-arranged in such a way
that the above condition holds.
The equations (5.111) are rewritten in the following form:
1
(b1 a12 x2 a13 x3 a1n xn )
a11
1
(b2 a21 x1 a23 x3 a2n xn )
x2 =
a22

1
(bn an1 x1 an2 x2 an n1 xn1 ).
xn =
ann
x1 =

(0)

(0)

(5.112)

(0)

To solve these equations an initial approximation x2 , x3 , . . . , xn for the variables x2 , x3 , . . . , xn respectively is considered. Substituting these values to the above
(1)
system and get the rst approximate value of x1 , denoted by x1 . Now, substitut(1)
(0) (0)
(0)
(1)
ing x1 for x1 and x3 , x4 , . . . , xn for x3 , x4 , . . . , xn respectively and we nd x2
from second equation of (5.112), the rst approximate value of x2 . Then substituting
(1) (1)
(1)
(0)
(0)
x1 , x2 , . . . , xi1 , xi+1 , . . . , xn for x1 , x2 , . . . , xi1 , xi+1 , . . . , xn to the ith equation of
(1)

## (5.112) respectively and obtain xi , and so on.

(k)
If xi , i = 1, 2, . . . , n be the kth approximate value of xi , then the (k + 1)th approximate value of x1 , x2 , . . . , xn are given by
1
(k)
(k)
(b1 a12 x2 a13 x3 a1n x(k)
n )
a11
1
(k+1)
(k+1)
(k)
=
(b2 a21 x1
a23 x3 a2n xn(k) )
(5.113)
x2
a22

1
(k+1)
(k+1)
(k+1)
(k)
(k)
=
(bi ai1 x1
ai i1 xi1 ai i+1 xi+1 an n1 xn1 )
xi
aii

1
(k+1)
(k+1)
(k+1)
=
(bn an1 x1
an2 x2
an n1 xn1 ).
x(k+1)
n
ann
k = 0, 1, 2, . . . .
(k+1)

x1

That is,
(k+1)
xi

i1
n



1 
(k+1)
(k)
=
aij xj

aij xj , i = 1, 2, . . . , n and k = 0, 1, 2, . . . .
bi
aii
j=1

j=i+1

(k+1)

(k)

## xi | < for all i = 1, 2, . . . , n, where > 0 is

The method is repeated until |xi
any pre-assigned number called the error tolerance. This method is called Gauss-Seidals
iteration method.

## 346 Numerical Analysis

Example 5.18.1 Solve the following system of equations by Gauss-Seidals iteration method, correct up to four decimal places.
27x + 6y z = 54
6x + 15y + 2z = 72
x + y + 54z = 110
Solution. The iteration scheme is
1
(54 6y (k) + z (k) )
27
1
= (72 6x(k+1) 2z (k) )
15
1
= (110 x(k+1) y (k+1) ).
54

x(k+1) =
y (k+1)
z (k+1)

Let y = 0, z = 0 be the initial solution. The successive iterations are shown below.
k
0
1
2
3
4
5
6

2.00000
1.18244
1.16735
1.16642
1.16635
1.16634

y
0
4.00000
4.07023
4.07442
4.07477
4.07480
4.07480

z
0
1.92593
1.93977
1.93997
1.93998
1.93998
1.93998

## The solution correct up to four decimal places is x = 1.1663, y = 4.0748, z = 1.9400.

Note 5.18.1 This solution is achieved in eleven iterations using Gauss-Jacobis method
while only six iterations are used in Gauss-Seidals method.
The sucient condition for convergence of this method is that the diagonal elements
of the coecient matrix are diagonally dominant. This is justied in the following.
5.18.1

## Convergence of Gauss-Seidals method



n
i1
1
1 
|aij | and let Ai =
|aij |, i = 1, 2, . . . , n.
Let A = max
i
aii j=1
aii
j=i

j=1

347

(k+1)
|
|i


1 
(k+1)
(k)

|aij | |j
|+
|aij | |j |
|aii |
j<i
j>i


1 
(k+1)
(k)
|aij | 
+
|aij |  

|aii |
j<i

Ai 

(k+1)

j>i

 + (A Ai )

(k)

.

## Then for some i,

(k+1)  Ai (k+1)  + (A Ai )(k) 
That is,
(k+1) 

A Ai (k)
 .
1 Ai

A Ai
A.
1 Ai
Therefore the above relation reduces to
Since 0 Ai A < 1 then

(k+1)  A(k) .

(5.114)

This shows that the rate of convergence of Gauss-Seidals iteration is also linear. The
successive substitutions give
(k)  Ak (0) .
Now, if A < 1 then (k)  0 as k , i.e., the sequence {x(k) } is sure to converge
when A < 1 i.e.,
n

|aij < |aii | for all i.
j=1
j=i

In other words the sucient condition for Gauss-Seidals iteration is that the coecient
matrix is diagonally dominant. The absolute error at the (k + 1)th iteration is given by
(k+1) 

A
e(k)  when A < 1,
1A

as in previous section.
Note 5.18.2 Usually, the Gauss-Seidals method converges rapidly than the GaussJacobis method. But, this is not always true. There are some examples in which the
Gauss-Jacobis method converges faster than the Gauss-Seidals method.

## Example 5.18.2 Solve the following system of equations by Gauss-Seidals method

correct to four signicant gures: 3x + y + z = 3, 2x + y + 5z = 5, x + 4y + z = 2.
Solution. It may be noted that the given system is not diagonally dominant, but,
the rearranged system 3x + y + z = 3, x + 4y + z = 2, 2x + y + 5z = 5 is diagonally
dominant.
Then the Gauss-Seidals iteration scheme is
1
x(k+1) = (3 y (k) z (k) )
3
1
y (k+1) = (2 x(k+1) z (k) )
4
1
(k+1)
= (5 2x(k+1) y (k+1) ).
z
5
Let y = 0, z = 0 be the initial values. The successive iterations are shown below.
k
0
1
2
3
4
5
6
7

1.00000
0.73333
0.71667
0.72139
0.72312
0.72340
0.72341

y
0
0.25000
0.17917
0.15313
0.14898
0.14881
0.14890
0.14893

z
0
0.55000
0.67083
0.68271
0.68165
0.68099
0.68086
0.68085

## Therefore, the solution correct up to four signicant gures is

x = 0.7234, y = 0.1489, z = 0.6808.
Example 5.18.3 Solve the following system of equations using Gauss-Seidals
method: 3x + y + 2z = 6, x + 4y + 2z = 5, 2x + y + 4z = 7.
Solution. The iteration scheme is
1
x(k+1) = (6 y (k) 2z (k) )
3
1
y (k+1) = (5 + x(k+1) 2z (k) )
4
1
(k+1)
= (7 2x(k+1) y (k+1) ).
z
4
Let y = 0, z = 0 be the initial solution and other approximate values are shown
below.

k
0
1
2
3
4
5
6
7
8
9
10

2.00000
1.20833
1.00347
0.89757
0.97866
0.99964
1.00163
1.00071
1.00016
1.00001

y
0
1.75000
1.39583
1.10243
0.92328
0.95946
0.98951
0.99901
1.00046
1.00028
1.00008

349

z
0
0.31250
0.79688
1.10243
1.07042
1.02081
1.00280
0.99943
0.99953
0.99985
0.99998

## Therefore, the solution correct up to four decimal places is

x = 1.0000, y = 1.0000, z = 1.0000.
It may be noted that the given system is not diagonally dominant while the iteration
scheme converges to the exact solution.
Another interesting problem is considered in the following. The system of equations
x1 + x2 = 2, x1 3x2 = 1
converges when the iteration scheme is taken as
1
(k+1)
(k)
(k+1)
(k+1)
= 2 x2 ,
x2
= (1 + x1
)
(5.115)
x1
3
While the Gauss-Seidals iteration method diverges when the iteration scheme is
(k+1)

x1

(k)

= 1 + 3x2 ,

(k+1)

x2

(k+1)

= 2 x1

(5.116)

## (It may be noted that the system is not diagonally dominant).

The calculations for these two schemes are shown below and the behaviour of the
solutions are shown in the gures 5.1 and 5.2.
For the scheme (5.115)
k x1
x2
0
0
1 2 0.333
2 1.667 0.222
3 1.778 0.259
4 1.741 0.247
5 1.753 0.251
6 1.749 0.250
7 1.750 0.250

For the
k
0
1
2
3
4
5
6
7

scheme (5.116)
x1 x2
0
1
1
4
-2
-5
7
22 -20
-59 61
184 -182
547 -545

## 350 Numerical Analysis

The exact solution of the equations is
x1 = 1.75, x2 = 0.25.
This example shows that the condition diagonally dominant is a sucient condition,
not necessary for Gauss-Seidals iteration method.
6

(0,2)

Second equ.

? 6

(1,0)

(2,0)
First equ.

(0,-1/3)
Figure 5.1: Illustration of Gauss-Seidals method for the convergent scheme (5.115.)
6
6

## (0,2) First equ.

(-5,0)

(2,0)


(4,0)
?

(0,-2)
Second equ.

Figure 5.2: Illustration of Gauss-Seidals method for the divergent scheme (5.116).
Algorithm 5.9 (Gauss-Seidals). This algorithm nds the solution of a system
of linear equations by Gauss-Seidals iteration method. The method will terminate
(k+1)
(k)
xi | < , where is the supplied error tolerance, for all i.
when |xi

## Solution of System of Linear Equations

Algorithm Gauss Seidal
Step 1. Read the coecients aij , i, j = 1, 2, . . . , n and the right hand vector
bi , i = 1, 2, . . . , n of the system of equations and error tolerance .
Step 2. Rearrange the given equations, if possible, such that the system becomes
diagonally dominant.
Step 3. Rewrite 
the ith equation as



1
bi
for i = 1, 2, . . . , n.
aij xj
aij xj ,
xi =
aii
j<i

j>i

## Step 4. Set the initial solution as

xi = 0, i = 1, 2, 3, . . . , n.
Step 5. Calculatethe new values xni of xi as


1
bi
xni =
aij xnj
aij xj ,
aii
j<i

for i = 1, 2, . . . , n.

j>i

Step 6. If |xi xni | < ( is an error tolerance) for all i then goto Step 7 else
set xi = xni for all i and goto Step 5.
Step 7. Print xni , i = 1, 2, . . . , n as solution.
end Gauss Seidal
Program 5.9
.
/* Program Gauss-Seidal
Solution of a system of linear equations by Gauss-Seidals
iteration method. Assume that the coefficient matrix satisfies
the condition of convergence. */
#include<stdio.h>
#include<math.h>
void main()
{
float a,b,x,xn,epp=0.00001,sum;
int i,j,n,flag;
printf("Enter number of variables ");
scanf("%d",&n);
printf("\nEnter the coefficients rowwise ");
for(i=1;i<=n;i++)
for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("\nEnter right hand vector ");
for(i=1;i<=n;i++)
scanf("%f",&b[i]);
for(i=1;i<=n;i++) x[i]=0; /* initialize */
/* testing of diagonal dominance may be included here
from the program of Gauss-Jacobis method */

351

## 352 Numerical Analysis

do
{
for(i=1;i<=n;i++)
{
sum=b[i];
for(j=1;j<=n;j++)
{
if(j<i)
sum-=a[i][j]*xn[j];
else if(j>i)
sum-=a[i][j]*x[j];
xn[i]=sum/a[i][i];
}
}
flag=0; /* indicates |x[i]-xn[i]|<epp for all i */
for(i=1;i<=n;i++) if(fabs(x[i]-xn[i])>epp) flag=1;
if(flag==1) for(i=1;i<=n;i++) x[i]=xn[i]; /* reset x[i] */
}
while(flag==1);
printf("Solution is \n");
for(i=1;i<=n;i++) printf("%8.5f ",xn[i]);
} /* main */
A sample of input/output:
Enter number of variables 3
Enter the coefficients rowwise
3 1 -1
2 5 2
2 4 6
Enter right hand vector
7 9 8
Solution is
2.00000 1.00000 0.00000

5.19

## The relaxation method is also an iterative method invented by Southwell in 1946.

(k) (k)
(k)
Let x(k) = (x1 , x2 , . . . , xn )t be the solution obtained at the kth iteration of the

## Solution of System of Linear Equations

353

system of equations
n


aij xj = bi , i = 1, 2, . . . , n.

(5.117)

j=1

Then

n


(k)

aij xj

= bi , i = 1, 2, . . . , n.

j=1
(k)

If ri

## denotes the residual of the ith equation, then

(k)

ri

= bi

n


(k)

aij xj .

(5.118)

j=1

Now, the solution can be improved successively by reducing the largest residual to
zero at that iteration.
To achieve the fast convergence of the method, the equations are rearranged in such
a way that the largest coecients in the equations appear on the diagonals. Now, the
largest residual (in magnitude) is determined and let it be rp . Then the value of the
rp
.
variable xp be increased by dxp where dxp =
app
In other words, xp is changed to xp + dxp to relax rp , i.e., to reduce rp to zero. Then
the new solution after this iteration becomes


(k) (k)
(k)
(k)
x(k+1) = x1 , x2 , . . . , xp1 , xp + dxp , xp+1 , . . . , xn(k) .
The method is repeated until all the residuals become zero or very small.
Example 5.19.1 Solve the following system of equations
2x1 + x2 + 9x3 = 12,
x1 7x2 + 2x3 = 4
8x1 + x2 x3 = 8,
by relaxation method taking (0, 0, 0) as initial solution.
Solution. We rearrange the equations to get the largest coecients in the diagonals
as
8x1 + x2 x3 = 8
x1 7x2 + 2x3 = 4
2x1 + x2 + 9x3 = 12.
The residuals r1 , r2 , r3 can be computed from the equations
r1 = 8 8x1 x2 + x3
r2 = 4 x1 + 7x2 2x3
r3 = 12 2x1 x2 9x3 .

## 354 Numerical Analysis

The method is started with the initial solution (0, 0, 0), i.e., x1 = x2 = x3 = 0. Then
the residuals r1 = 8, r2 = 4, r3 = 12 of which the largest residual in magnitude
is r3 . This indicates that the third equation has more error and needs to improve
r3
12
= 1.333. The new solution then becomes
=
x3 . Then the increment dx3 =
a33
9
(0, 0, 0 + 1.333) i.e., (0, 0, 1.333).
Similarly, we nd the new residuals of large magnitudes and relax it to zero and
we repeat the process until all the residuals become zero or very small.
The detail calculations are shown in the following table.
k
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14

r1

8
9.333
0.003
1.122
1.505
0.001
0.135
0.192
0
0.019
0.027
0.003
0.001
0

residuals
max
r2
r3
(r1 , r2 , r3 )

4
12
12
6.666 0.003
9.333
7.833 2.331 7.833
0
3.450 3.450
0.766 0.003 1.505
0.954 0.373
0.954
0.002 0.509
0.509
0.112 0.004
0.192
0.136 0.052 0.136
0.003 0.071 0.071
0.013 0.001
0.027
0.016 0.007
0.016
0.002 0.009
0.009
0
0
0

3
1
2
3
1
2
3
1
2
3
1
2
3

increment
dxp

1.333
1.167
1.119
0.383
0.188
0.136
0.057
0.024
0.019
0.008
0.003
0.002
0.001
0

solution
x1
x2
x3
0
0
0
0
0 1.333
1.167 0 1.333
1.167 1.119 1.333
1.167 1.119 0.950
0.979 1.119 0.950
0.979 0.983 0.950
0.979 0.983 1.007
1.003 0.983 1.007
1.003 1.002 1.007
1.003 1.002 0.999
1.000 1.002 0.999
1.000 1.000 0.999
1.000 1.000 1.000
1.000 1.000 1.000

At this stage all the residuals are zero and therefore the solution of the given system
of equations is x1 = 1.000, x2 = 1.000, x3 = 1.000, which is the exact solution of the
equations.

5.20

## The convergence can be accelerated by introducing a suitable relaxation factor w. In

n

aij xj = bi can be written as
this method, the ith equation
j=1
i1

j=1

aij xj +

n

j=i

aij xj = bi .

(5.119)

355

## Like Gauss-Seidals method, for the solution



(k+1) (k+1)
(k+1) (k) (k)
,
x1
, x2
, . . . , xi1 , xi , xi+1 , . . . , x(k)
n
the equation (5.119) becomes
i1


(k+1)

aij xj

j=1

n


(k)

aij xj

= bi .

(5.120)

j=i

ri = bi

i1


(k+1)
aij xj

j=1
(k)

(k+1)

n


(k)

aij xj .

(5.121)

j=i

(k)

## xi denote the dierences of xi at two consecutive iterations.

Let di = xi
In the successive overrelaxation (S.O.R. or SOR) method, assume that
(k)

aii di

= wri , i = 1, 2, . . . , n,

(5.122)

## where w is a suitable factor, called the relaxation factor.

Using (5.121), equation (5.122) becomes
(k+1)
aii xi

(k)
aii xi


i1
j=1

(k+1)
aij xj

n


(k)
aij xj

bi ,

(5.123)

j=i

i = 1, 2, . . . , n; k = 0, 1, 2, . . .
6 (0) (0)
(0) 7t
and x1 , x2 , . . . , xn
is the initial solution. The method is repeated until desired
accuracy is achieved.
This method is called the overrelaxation method when 1 < w < 2, and is called
the under relaxation method when 0 < w < 1. When w = 1, the method becomes
Gauss-Seidals method.
Best relaxation factor wb
The proper choice of wb can speed up the convergence of the system. In a problem,
Carre took wb = 1.9 and found that the convergence is 40 times faster than when w = 1
(Gauss-Seidals method). He also observed that when w = 1.875 (a minor variation
of 1.9), the convergence is only two times faster than the Gauss-Seidals method. In
general, the choice of wb is not a simple task. It depends on the spectral radius of the
coecient matrix.

## Example 5.20.1 Solve the following system of equations

3x1 + x2 + 2x3 = 6
x1 + 4x2 + 2x3 = 5
2x1 + x2 + 4x3 = 7
by SOR method taken w = 1.01
Solution. The iteration scheme for SOR method is


(k+1)
(k)
(k)
(k)
(k)
a11 x1
= a11 x1 w a11 x1 + a12 x2 + a13 x3 b1


(k+1)
(k)
(k+1)
(k)
(k)
= a22 x2 w a21 x1
+ a22 x2 + a23 x3 b2
a22 x2


(k+1)
(k)
(k+1)
(k+1)
(k)
= a33 x3 w a31 x1
+ a32 x2
+ a33 x3 b3
a33 x3
or
(k+1)

3x1

(k+1)

4x2

(k+1)

4x3
(0)

(0)



(k)
(k)
(k)
(k)
= 3x1 1.01 3x1 + x2 + 2x3 6


(k)
(k+1)
(k)
(k)
= 4x2 1.01 x1
+ 4x2 + 2x3 5


(k)
(k+1)
(k+1)
(k)
= 4x3 1.01 2x1
+ x2
+ 4x3 7 .

(0)

Let x1 = x2 = x3 = 0.
The detail calculations are shown in the following table.
k
0
1
2
3
4
5
6
7
8
9

x1
0
2.02000
1.20116
0.99557
0.98169
0.99312
0.99879
1.00009
1.00013
1.00005

x2
0
1.77255
1.39665
1.09326
1.00422
0.99399
0.99728
0.99942
0.99999
1.00005

x3
0
0.29983
0.80526
0.98064
1.00838
1.00491
1.00125
1.00009
0.99993
0.99997

## Therefore the required solution is

x1 = 1.0000, x2 = 1.0000, x3 = 1.0000
correct up to four decimal places.

## Solution of System of Linear Equations

Program 5.10
.
/* Program Gauss-Seidal SOR
Solution of a system of linear equations by Gauss-Seidal
successive overrelaxation (SOR) method. The relaxation factor w
lies between 1 and 2. Assume that the coefficient matrix
satisfies the condition of convergence. */
#include<stdio.h>
#include<math.h>
void main()
{
float a,b,x,xn,epp=0.00001,sum,w;
int i,j,n,flag;
printf("Enter number of variables ");
scanf("%d",&n);
printf("\nEnter the coefficients rowwise ");
for(i=1;i<=n;i++)
for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("\nEnter right hand vector ");
for(i=1;i<=n;i++) scanf("%f",&b[i]);
printf("Enter the relaxation factor w ");
scanf("%f",&w);
for(i=1;i<=n;i++) x[i]=0; /* initialize */
for(i=1;i<=n;i++) printf(" x[%d] ",i);printf("\n");
do
{
for(i=1;i<=n;i++)
{
sum=b[i]*w+a[i][i]*x[i];
for(j=1;j<=n;j++)
{
if(j<i)
sum-=a[i][j]*xn[j]*w;
else if(j>=i)
sum-=a[i][j]*x[j]*w;
xn[i]=sum/a[i][i];
}
}
for(i=1;i<=n;i++) printf("%8.5f ",xn[i]);
printf("\n");

357

## flag=0; /* indicates |x[i]-xn[i]|<epp for all i */

for(i=1;i<=n;i++) if(fabs(x[i]-xn[i])>epp) flag=1;
if(flag==1) for(i=1;i<=n;i++) x[i]=xn[i]; /* reset x[i] */
}while(flag==1);
printf("Solution is \n");
for(i=1;i<=n;i++) printf("%8.5f ",xn[i]);
} /* main */
A sample of input/output:
Enter number of variables 3
Enter the coefficients rowwise
9 2 4
1 10 4
2 -4 10
Enter right hand vector
20 6 -15
Enter the relaxation factor w 1.01
x
x
x
2.24444 0.37931 -1.81514
2.95166 1.03740 -1.67397
2.73352 0.99583 -1.64812
2.73342 0.98581 -1.65241
2.73760 0.98722 -1.65264
2.73734 0.98732 -1.65254
2.73728 0.98729 -1.65254
2.73729 0.98729 -1.65254
Solution is
2.73729 0.98729 -1.65254

5.21

## Comparison of Direct and Iterative Methods

The direct and iterative, both the methods have some advantages and also some disadvantages and a choice between them is based on the given system of equations.
(i) The direct method is applicable for all types of problems (when the coecient
determinant is not zero) where as iterative methods are useful only for particular
types of problems.
(ii) The rounding errors may become large particularly for ill-conditioned systems
while in iterative method the rounding error is small, since it is committed in

## Solution of System of Linear Equations

359

the last iteration. Thus for ill-conditioned systems an iterative method is a good
choice.
(iii) In each iteration, the computational eect is large in direct method (it is 2n3 /3
for elimination method) and it is low in iteration method (2n2 in Gauss-Jacobis
and Gauss-Seidals methods).
(iv) Most of the direct methods are applied on the coecient matrix and for this
purpose, the entire matrix to be stored into primary memory of the computer.
But, the iteration methods are applied in a single equation at a time, and hence
only a single equation is to be stored at a time in primary memory. Thus iterative
methods are ecient then direct method with respect to space.

5.22

Exercise




 2 3
0 2 5



 6 1
 1 3 8 

(i)
(ii)


 8 3
6 5 1

 3 8

pivoting.

8 4 
0 5 
.
1 2 
7 10 

## 2. Solve the following systems of equations by Cramers rule.

(ii) 7.6x1 + 0.5x2 + 2.4x3 = 1.9
(i) 3x1 + 2x2 + x3 = 5
2x1 + 5x2 + x3 = 3
2.2x1 + 9.1x2 + 4.4x3 = 9.7
2x1 + x2 + 3x3 = 11
1.3x1 + 0.2x2 + 5.8x3 = 1.4
3. Find the inverse of the matrix

11 3 1
2 5 5
1 1 1

## using Gauss-Jordan method and solve the following system of equations.

11x1 + 3x2 x3 = 15
2x1 + 5x2 + 5x3 = 11
x1 + x2 + x3 = 1.
4. Find the inverses of the following matrices (using partial pivoting).

1 2 0
0 1 2
1 0 5 .
3 5 1
(ii)
(i)
3 8 7
6 8 9

## 360 Numerical Analysis

5. Solve the following systems of equations by Gauss elimination method.
(ii)
4x + y + z = 4
(i)
x1 + 12 x2 + 13 x3 = 1
1
1
1
x + 4y 2z = 4
2 x1 + 3 x2 + 4 x3 = 0
1
1
1
x
+
x
+
x
=
0
3x + 2y 4z = 6
3 1
4 2
5 3
= 6
(iii) x1 4x2 x4
(iv) 1.14x1 2.15x2 5.11x3 = 2.05
x1 + x2 + 2x3 + 3x4 = 1
0.42x1 1.13x2 + 7.05x3 = 0.80
2x1 + 3x2 x3 x4 = 1
0.71x1 + 0.81x2 0.02x3 = 1.07
x1 + 2x2 + 3x3 x4 = 3
6. Use Gauss elimination method to nd the values of the determinants.



1 4 1 3
 1.6 5.4 7.7 3.1



 0 1 3 1 
 8.2 1.4 2.3 0.2



(i)
(ii)
3 1 0 2
 5.3 5.9 2.7 8.9



 1 2 5 1 
 0.7 1.9 8.5 4.8





.




## 7. Using LU decomposition method, solve the following systems of equations

(ii)
x 2y + 7z = 6
(i)
x1 + x2 + x3 = 3
4x + 2y + z = 7
2x1 x2 + 3x3 = 16
2x + 5y 2z = 5.
3x1 + x2 x3 = 3
8. Find the triangular factorization A = LU for the matrices

4 2 1
5 2 1
2 5 2
1 0 3 .
(i)
(ii)
1 2 7
3 1 6
9. Solve LY = B, UX = Y
A = LU is given by

4 8 4 0
1 5 4 3

1 4 7 2
1 3 0 2

1
0
0
1/4 1
0
=
1/4 2/3
1
1/4 1/3 1/2

0
4 8

0 0 3
0 0 0
1
0 0

4 0
3 3
.
4 4
0 1

## 10. Prove that the product of two upper-triangular matrices is an upper-triangular

matrix.
11. Prove that the inverse of a non-singular upper-triangular matrix is an uppertriangular matrix.
12. Use Cholesky method to solve the following systems of linear equations.
(ii)
x1 + 3x2 + 4x3 = 8
(i)
x + 2y + 3z = 0
3x1 x2 + 5x3 = 7
2x + y + 2z = 1
3x + 2y + z = 4
4x1 + 5x2 7x3 = 2.

## 13. Find the inverses of the following matrices using partition.

20 1 2
8 1 1
3 20 1
2 1 9 .
(i)
(ii)
2 3 20
1 7 2
14. Find the solution of the following tri-diagonal system:
2x1 2x2 = 1
x1 + 2x2 3x3 = 2
2x2 + 2x3 4x4 = 1
x3 x4 = 3.
15. Test the following system for ill-condition.
10x + 7y + 8z + 7w = 32
7x + 5y + 6z + 5w = 23
8x + 6y + 10z + 9w = 33
7x + 5y + 9z + 10w = 31.
16. Find the g-inverses of the following matrices

(i)

2 3
4 6

(ii)

2 3 5
1 1 0 .
3 1 2

1 1 1
2 3 5

## and hence solve the following system of equations.

x+y+z = 3
2x + 3y + 5z = 10.
18. Find the least squares solution of the equations
x1 + 2x2 = 3
.
2x1 + 4x2 = 7

361

## 362 Numerical Analysis

19. Find the solution of the following equations using least squares method:
x + y + 3z = 0
2x y + 4z = 8
x + 5y + z = 10
x + y 2z = 2.
20. Solve the following equations by (i) Gauss-Jordans and (ii) Gauss-Seidals methods, correct up to four signicant gures:
9x + 2y + 4z = 20
x + 10y + 4z = 6
2x 4y + 10z = 15.
21. Test if the following systems of equations are diagonally dominant and hence solve
them using Gauss-Jacobis and Gauss-Seidals methods.
(i)
10x + 15y + 3z = 14
(ii)
x + 3y + 4z = 7
3x + 2y + 5z = 10
30x + y + 5z = 17
x 5y + 7z = 3.
x + y + 4z
= 3
22. Solve the following simultaneous equations
2.5x1 + 5.2x2 = 6.2
1.251x1 + 2.605x2 = 3.152
by Gauss elimination method and get your answer to 4 signicant gures. Improve
the solution by iterative renement.
23. Consider the following linear system
2x1 x2 = 4.
5x1 + 3x2 = 6,
Can either Gauss-Jacobis or Gauss-Seidals iteration be used to solve this linear
system? Why?
24. Consider the following tri-diagonal linear system and assume that the coecient
matrix is diagonally dominant.
d1 x1 + c1 x2
a1 x1 + d2 x2 +
a2 x2 +
..
.

c2 x3
d3 x3
..
.
an2 xn2

=
=
=

b1
b2
b3
..
.

c4 x4
..
.
+ dn1 xn1 + cn1 xn = bn1
an1 xn1 + dn xn = bn .

363

## 25. Use Gauss-Seidals iteration to solve the following band system.

12x1 2x2 + x3 = 5
2x1 + 12x2 2x3 + x4 = 5
x1 2x2 + 12x3 2x4 + x5 = 5
x2 2x3 + 12x4 2x5 + x6 = 5

x46 2x47 + 12x48 2x49 + x50 = 5
x47 2x48 + 12x49 2x50 = 5
x48 2x49 + 12x50 = 5.
26. Solve the following systems of equations by successive relaxation method.
(ii)
x+yz
= 0
(i)
2x 3y + z = 1
2x + 3y + 8z = 1
x + 4y 3z = 0
5x 4y + 10z = 9.
x+y+z
= 6
27. Solve the following systems of equations by successive overrelaxation method.
(ii)
x1 + x2 + x3
(i)
5x 2y + z = 4
=
3
7x + y 5z = 8
2x1 x2 + 3x3 = 16
3x + 7y + 4z = 10
3x1 + x2 x3 = 3.
Choose appropriate relaxation factor w.
28. The Hilbert matrix is a classical ill-conditioned matrix, and small changes in its
coecients will produce a large changes in the solution to the perturbed system.
(i) Solve AX = b using the Hilbert matrix

1 1/2 1/3 1/4 1/5
1
1/2 1/3 1/4 1/5 1/6
0

, b = 0 .
1/3
1/4
1/5
1/6
1/7
A=

1/4 1/5 1/6 1/7 1/8
0
1/5 1/6 1/7 1/8 1/9
0
(ii) Solve Cx = b where

1.0
0.5
0.33333 0.25
0.2
0.5
0.33333 0.25
0.2
0.16667

0.33333
0.25
0.2
0.16667
0.14286
C=

0.25
0.2
0.16667 0.14286 0.125
0.2
0.16667 0.14286 0.125 0.11111,
[Note that the two matrices A and C are dierent.]

, b =

1
0
0
0
0

Chapter 6

## Eigenvalues and Eigenvectors

of a Matrix
6.1

Eigenvalue of a Matrix

such that
AX = X.

(6.1)

## The scalar is called an eigenvalue or characteristic value of the matrix A

and X is called the corresponding eigenvector. The equation (6.1) can be written as
(A I)X = 0.
The equation
|A I| = 0

(6.2)

that is,

 a11 a12

 a21 a22




 an1
an2

a13
a23

an3


a1n 
a2n 
=0


ann 

(6.3)

## is a polynomial in of degree n, called characteristic equation of the matrix A. The

roots i , i = 1, 2, . . . , n, of the equation (6.2) are the eigenvalues of A. For each value
of i , there exists an Xi such that
AXi = i Xi .
365

(6.4)

## 366 Numerical Analysis

The eigenvalues i may be either distinct or repeated, they may be real or complex.
If the matrix is real symmetric then all the eigenvalues are real. If the matrix is skewsymmetric then the eigenvalues are either zero or purely imaginary. Sometimes, the set
of all eigenvalues, i , of a matrix A is called the spectrum of A and the largest value
of |i | is called the spectral radius of A.
Example 6.1.1 Find the eigenvalues and eigenvectors of the matrix

2 1 1
A = 1 2 1 .
1 1 2
Solution. The characteristic equation of A is |A I| = 0.
Therefore


 2 1
1 

 1 2 1  = 0.


 1
1 2 
By direct expansion this gives (1 )2 (4 ) = 0.
Hence the characteristic equation is (1 )2 (4 ) = 0 and the eigenvalues of A are
1, 1, and 4. The two distinct eigenvectors corresponding to two eigenvalues = 1
and 4 are calculated below.
Eigenvector corresponding to the eigenvalue 1.
Let (x1 , x2 , x3 )T be the eigenvector corresponding to 1. Then

0
2 1 1
1
x1
1 2 1 1 x2 = 0 .
x3
0
1
1 2 1
Thus
x1 x2 + x3 = 0
x1 + x2 x3 = 0
x1 x2 + x3 = 0.
The solution of this system of equations is x3 = 0, x1 = x2 . We take x2 = 1. Then
the eigenvector is (1, 1, 0)T .
Let (y1 , y2 , y3 )T be the eigenvector corresponding to = 4.
Then

0
2 4 1
1
y1
1 2 4 1 y2 = 0
y3
0
1
1 2 4

367

## Thus the system of equations are

2y1 y2 + y3 = 0
y1 2y2 y3 = 0
y1 y2 2y3 = 0.
The solution is y1 = k, y2 = k, y3 = k for arbitrary k. Let k = 1. Then the
eigenvector is (1, 1, 1)T .
The upper bound of the eigenvalue of a matrix is stated below.
Theorem 6.1 The largest eigenvalue (in magnitude) of a square matrix A cannot exceed the largest sum of the moduli of the elements along any row or any column, i.e.,
n
n




|| max
|aij | and || max
|aij | .
(6.5)
i

j=1

i=1

## Proof. Let be an eigenvalue of A and X = (x1 , x2 , . . . , xn )T be the corresponding

eigenvector. Then AX =X.
This equation may be written as
a11 x1 + a12 x2 + + a1n xn = x1
a21 x1 + a22 x2 + + a2n xn = x2

ak1 x1 + ak2 x2 + + akn xn = xk

an1 x1 + an2 x2 + + ann xn = xn .
Let |xk | = max |xi |. Now, the kth equation is divided by xk . Therefore,
i

= ak1

x1
x2
xn
+ ak2
+ + akk + + akn .
xk
xk
xk

x 
 i
Since   1 for i = 1, 2, . . . , n,
xk

n


|akj |.

j=1

## This is true for all rows k, hence

|| max
i

n



|aij | .

j=1

The theorem is also true for columns, as the eigenvalues of AT and A are same. 

## 368 Numerical Analysis

Theorem 6.2 (Shifting eigenvalues). Suppose be an eigenvalue and X be its
corresponding eigenvector of A. If c is any constant, then c is an eigenvalue of the
matrix A cI with same eigenvector X.
Let the characteristic polynomial of the matrix A be
det(A I) = n + c1 n1 + c2 n2 + + cn ,

(6.6)

where
c1 =

n


aii = T r. A, which is the sum of all diagonal elements of A, called the trace.

i=1



  aii aij 


c2 =
 aji ajj  ,
i<j

  aii aij
 aji ajj
c3 =

i<j<k  aki akj

## is the sum of all principal minors of order two of A,

aik
ajk
akk




 , is the sum of all principal minors of order three of A,



## and nally (1)n cn = detA, is the determinant of the matrix A.

If the coecients of the polynomial (6.6) are known then solving this polynomial
using the method discussed in Chapter 4, one can determine all the eigenvalues. But,
the direct expansion is very labourious and is only used for low order matrices. One
ecient method to compute the coecients of (6.6) is Leverrier-Faddeev method.

6.2

## This method was developed by Leverrier and latter modied by Faddeev.

Let
det(A I) = n + c1 n1 + c2 n2 + + cn

(6.7)

## be the characteristic polynomial of the matrix A and its roots be 1 , 2 , . . . , n . Now,

the sum of kth power of the eigenvalues is denoted by Sk , i.e.,
S1
S2

Sn

=
=

1 + 2 + + n = T r A,
21 + 22 + + 2n = T r A2 ,

n1 + n2 + + nn = T r An .

(6.8)

## Then by Newtons formula (on polynomial) for k n

Sk + c1 Sk1 + c2 Sk2 + + ck1 S1 = kck .

(6.9)

369

## For k = 1, 2, . . . , n, the coecients are given by

c1 = S1
1
c2 = (S2 + c1 S1 )
2

1
cn = (Sn + c1 Sn1 + c2 Sn2 + + cn1 S1 ).
n
Thus computation of the coecients c1 , c2 , . . . , cn depends on the trace of A, A2 , . . .,
An .
The Leverrier method was modied by Faddeev by generating a sequence of matrices
B1 , B2 , . . . , Bn , shown below.
B1
B2
B3

Bn1
Bn

=
=
=

=
=

A,

d1 =
T r. B1 ,
1
d2 =
2 T r. B2 ,
1
d3 =
3 T r. B3 ,

1
dn1 = n1
T r. Bn1 ,
1
dn =
T
r.
Bn ,
n

D1 =
D2 =
D3 =

Dn1 =
Dn =

B1 d 1 I
B2 d 2 I
B3 d 3 I
(6.10)

Bn1 dn1 I
Bn d n I

## Thus the coecients of the characteristic polynomial are

c1 = d1 , c2 = d2 , . . . , cn = dn .
It can be veried that Dn = 0.
method to nd the characteristic poly
2 1 1
nomial of the matrix 0 2 3 .
1 5 4
Solution.

2 1 1
B1 = A = 0 2 3 , d1 = T r. B1 = 2 + 2 + 4 = 8
1 5 4

2 1 1
6 1 1
13 9 5
3 3 6
B2 = A(B1 d1 I) = 0 2 3 0 6 3 =
1 5 4
1 5 4
2 9 2
1
1
d2 = T r. B2 = (13 + 3 2) = 6
2
2

2 1 1
7 9 5
9 0 0
B3 = A(B2 d2 I) = 0 2 3 3 9 6 = 0 9 0
1 5 4
2 9 4
0 0 9

## 370 Numerical Analysis

d3 =

1
1
T r. B3 = (9 9 9) = 9.
3
3

## Thus c1 = d1 = 8, c2 = d2 = 6, c3 = d3 = 9. Hence the characteristic polynomial is 3 82 + 6 + 9 = 0.

Note 6.2.1 Using this method one can compute the inverse of the matrix A. It is
mentioned that Dn = 0. That is, Bn dn I = 0 or, ADn1 = dn I. From this relation
one can write Dn1 = dn A1 . This gives,
A1 =

Dn1
Dn1
=
.
dn
cn

(6.11)

## Algorithm 6.1 (Leverrier-Faddeev method). This algorithm determines the

characteristic polynomial n + c1 n1 + + cn1 + cn = 0 of a square matrix A.
Step 1: Read the matrix A = [aij ], i, j = 1, 2, . . . , n.
n

Step 2: Set B1 = A, d1 =
aii .
i=1

Step 3: for i = 2, 3, . . . , n do
Compute
(a) Bi = A(Bi1 di1 I)
(b) di = 1i (sum of the diagonal elements of Bi )
Step 4: Compute ci = di for i = 1, 2, . . . , n.
//ci s are the coecients of the polynomial.//
Program 6.1
.
This program finds the characteristic polynomial of
a square matrix. From which we can determine all the
eigenvalues of the matrix. */
#include<stdio.h>
#include<math.h>
void main()
{
int n,i,j,k,l;
float a,b,c,d;
printf("Enter the size of the matrix ");
scanf("%d",&n);

## Eigenvalues and Eigenvectors of a Matrix

printf("Enter the elements row wise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("The given matrix is\n");
for(i=1;i<=n;i++) /* printing of A */
{
for(j=1;j<=n;j++) printf("%f\t ",a[i][j]); printf("\n");
}
printf("\n");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) b[i][j]=a[i][j];
d=0;for(i=1;i<=n;i++) d+=a[i][i];
for(i=2;i<=n;i++)
{
/* construction of B[i-1]-d[i-1] I=C (say) */
for(j=1;j<=n;j++) for(k=1;k<=n;k++) c[j][k]=b[j][k];
for(j=1;j<=n;j++) c[j][j]=c[j][j]-d[i-1];
/* product of A and B[i-1]-d[i-1]I */
for(j=1;j<=n;j++) for(k=1;k<=n;k++)
{
b[j][k]=0;
for(l=1;l<=n;l++) b[j][k]+=a[j][l]*c[l][k];
}
/* trace of B */
d[i]=0;
for(j=1;j<=n;j++)
d[i]+=b[j][j];
d[i]/=i;
} /* end of i loop */
printf("The coefficients of the characteristic polynomial are \n");
printf("1.00000 ");
for(i=1;i<=n;i++) printf("%8.5f ",-d[i]);
}/* main */
A sample of input/output:
Enter
Enter
2 3 0
3 5 6
2 3 0
0 0 2
1 2 3

## the size of the matrix 5

the elements row wise
1 2
2 1
1 0
3 8
0 2

371

## 372 Numerical Analysis

The given matrix is
2.000000
3.000000
3.000000
5.000000
2.000000
3.000000
0.000000
0.000000
1.000000
2.000000

0.000000
6.000000
0.000000
2.000000
3.000000

1.000000
2.000000
1.000000
3.000000
0.000000

2.000000
1.000000
0.000000
8.000000
2.000000

## The coefficients of the characteristic polynomial are

1.00000 -12.00000 18.00000 -29.00000 -142.00000 14.00000
6.2.1

The Leverrier-Faddeev method may also be used to determine all the eigenvectors.
Suppose the matrices D1 , D2 , . . . , Dn1 and the eigenvalues 1 , 2 , . . . , n are known.
Then the eigenvectors x(i) can be determined using the formula
e2 + + en1 ,
x(i) = in1 e0 + in2 e1 + n3
i

(6.12)

where e0 is a unit vector and e1 , e2 , . . . , en1 are column vectors of the matrices
D1 , D2 , . . . , Dn1 of the same order as e0 .
methodto nd characteristic equation and

9 1 9
all eigenvectors of the matrix A = 3 1 3 .
7 1 7
Solution.
B1 =
d1 =
D1 =

B2 =
d2 =
D2 =

9 1 9
A = 3 1 3
7 1 7
T r. B1 = 9 1 7 = 1

9 1 9
1

3 1 3 0
B1 d 1 I =
7 1 7
0

9 1 9
8 1
AD1 = 3 1 3 3 2
7 1 7
7 1
1
1
T r. B2 = (6 + 2 4) = 2
2
2

4 2 6
B2 d 2 I = 0 0 0
4 2 6

0 0
8 1 9
1 0 = 3 2 3
0 1
7 1 8

9
6 2 6
3= 0 2 0
8
4 2 4

## Eigenvalues and Eigenvectors of a Matrix

373

9 1 9
4 2 6
0 0 0
B3 = AD2 = 3 1 3 0 0 0 = 0 0 0
7 1 7
4 2 6
0 0 0
1
1
d3 = T r. B3 = (0 + 0 + 0) = 0.
3
3
Thus c1 = d1 = 1, c2 = d2 = 2, c3 = d3 = 0.
The characteristic equation is 3 2 2 = 0.
The eigenvalues are 1 = 0, 2 = 1, 3 = 2.

1
8
4
Let e0 = 0 , and then e1 = 3 , e2 = 0 .
0
7
4
(e1 , e2 are the rst columns of the matrices D1 , D2 ).
The formula
x(i) = 2i e0 + i e1 + e2 ,

4
for 1 = 0 gives x(1) = 0 .
4
Similarly, for 2 = 1,

1
8
4
3
x(2) = (1)2 0 + (1) 3 + 0 = 3
0
7
4
3

1
8
4
24
6 .
and for 2 = 2, x(3) = 22 0 + 2 3 + 0 =
0
7
4
18

4
3
24
6 .
Thus the eigenvectors are 0 , 3 and
4
3
18
A square matrix B is said to be similar to another square matrix A if there exists
a non-singular matrix P such that B = P1 AP. The similar matrices have the same
eigenvalues. A square matrix A is said to be diagonalisable if A is similar to a square
diagonal matrix. A square matrix A of order n is diagonalisable i A has n linearly
independent eigenvectors. If P1 AP is a diagonal matrix and P is orthogonal then
A is said to be orthogonally diagonalisable. It can be proved that a matrix A is
orthogonally diagonalisable i A is real and symmetric.
If a matrix is either diagonal or lower triangular or upper triangular then its eigenvalues are the diagonal elements.

6.3

## Eigenvalues for Arbitrary Matrices

Several methods are available to determine the eigenvalues and eigenvectors of a matrix.
Here, two methods Rutishauser and Power methods are introduced.
6.3.1

Rutishauser method

## Let A be a square matrix. In this method, a convergent sequence of upper triangular

matrices A1 , A2 , . . . are generated. The diagonal elements of the convergent matrix are
the eigenvalues, if they are real.
The conversion is based on LU decomposition technique, where L and U are lower
and upper triangular matrices. Initially, let A = A1 and A1 = L1 U1 with lii = 1. Then
form A2 = U1 L1 . The matrices A1 and A2 are similar as A2 = U1 L1 = U1 A1 U1
1
and they have same eigenvalues. Now, A2 is factorized in the form A2 = L2 U2 with
lii = 1. Then form A3 = U2 L2 . Proceeding this way we generate a sequence of similar
matrices A1 , A2 , A3 , . . .. In general, this sequence converge to an upper triangular
matrix or a near-triangular matrix A. If the eigenvalues are real, then they all lie on
the leading diagonal of the matrix A. But, practically this method is complicated.
Sometimes, the lower triangular matrix L is replaced by Q, where Q is an unitary or
orthogonal matrix. The QU decomposition is also not simple for practical computation.
Since the sequence {Ai } converges slowly the shifting technique may be used to
accelerate its convergence. This technique is not discussed here.

4 2
Example 6.3.1 Find all the eigenvalues of the matrix
using Rutishauser
1 1
method.

u11 u12
1 0
.
Solution. Let A = A1 =
0 u22
l21 1

u12
4 2
u11
.
That is,
=
u11 l21 u12 l21 + u22
1 1
This gives u11 = 4, u12 = 2,
l21 = 1/4,
u22 =

3/2.
1 0
4 2
, U1 =
.
Therefore, L1 =
1/4 1
0 3/2

4 2
1 0
7/2 2
=
.
Form A2 = U1 L1 =
0 3/2
1/4 1
3/8 3/2

u11 u12
1 0
.
Again, let A2 = L2 U2 =
0
u22
l21 1

u12
7/2
2
u11
.
=
u11 l21 u12 l21 + u22
3/8 3/2

375

## Solution is u11 = 7/2, u12 = 2, l21 = 3/28, u22 = 12/7.

1
0
7/2 2
, U2 =
.
Therefore, L2 =
3/28 1
0 12/7

7/2 2
1
0
23/7
2
=
.
Form A3 = U2 L2 =
0 12/7
3/28 1
9/49 12/7
In this way, we nd

3.17391
2
3.10959
2
A4 =
, A5 =
,
0.10208 1.82609
0.06080 1.89041

3.07049
2
3.04591
2
, A7 =
,
A6 =
0.03772 1.92951
0.02402 1.95409

3.04591
2
3.04073
2
, A9 =
A8 =
0.00788 1.96986
0.00512 1.97504
and so on.
The sequence {Ai } converges slowly to an upper triangular matrix and the diagonal
elements converge to the eigenvalues of A. The exact eigenvalues are 3 and 2, which
are approximated by the diagonal elements of A9 .
6.3.2

Power method

## Power method is generally used to nd the eigenvalue, largest in magnitude, (sometimes

called rst eigenvalue) of a matrix A. Let 1 , 2 , . . . , n be all eigenvalues of the matrix
A. We assume that
|1 | > |2 | > |3 | > > |n |,
i.e., 1 is largest in magnitude and X1 , X2 , . . . , Xn be the eigenvectors corresponding
to the eigenvalues 1 , 2 , . . . , n respectively. The method is applicable if the matrix A
has n independent eigenvectors. Then any vector X in the (vector) space of eigenvectors
X1 , X2 , . . . , Xn can be written as
X = c1 X 1 + c2 X 2 + + c n X n .

(6.13)

## Multiplying this relation by A and using the results AX1 = 1 X1 , AX2 = 2 X2 ,

. . . , AXn = n Xn , we obtain
AX = c1 1 X1 + c2 2 X2 + + cn n Xn

 
 

2
n
X2 + + c n
Xn .
= 1 c1 X 1 + c 2
1
1

(6.14)

## 376 Numerical Analysis

A X =
..
.
Ak X =
Ak+1 X =

 2
 2

2
n
c1 X 1 + c2
X2 + + c n
Xn .
1
1
..
.

 k
 k

2
n
k
1 c1 X 1 + c2
X2 + + c n
Xn .
1
1

 k+1
 k+1

2
n
k+1
c1 X 1 + c2
1
X2 + + c n
Xn .
1
1
21

(6.15)

(6.16)
(6.17)

When k , then
right hand sides of (6.16) and (6.17) tend to k1 c1 X1 and



 i  < 1 for i = 2, . . . , n. Thus for k , Ak X = k c1 X1 and
k+1
1
1 c1 X1 , since 
1 

k
k+1 X. It is well known that two
Ak+1 X = k+1
1 c1 X1 . That is, for k , 1 A = A
vectors are equal if their corresponding components are same. That is,


Ak+1 X
 r , r = 1, 2, . . . , n.
1 = lim 
(6.18)
k
k
A X
r

The symbol (Ak X)r denotes the rth component of the vector Ak X.
If |2 |  |1 |, then the term within square bracket of (6.17) tend faster to c1 X1 , i.e.,
the rate of convergence is fast.
To reduce the round o error, the method is carried out by normalizing (reducing the
largest element to unity) the eigenvector at each iteration. Let X0 be a non-null initial
(arbitrary) vector (non-orthogonal to X1 ) and we compute
Yi+1 = AXi
Yi+1
Xi+1 = (i+1) ,

for i = 0, 1, 2, . . . .

(6.19)

where (i+1) is the largest element in magnitude of Yi+1 and it is the (i + 1)th approximate value of 1 . Then
(Yk+1 )r
,
r = 1, 2, . . . , n.
(6.20)
1 = lim
k (Xk )r
and Xk+1 is the eigenvector corresponding to the eigenvalue 1 .
Note 6.3.1 The initial vector X0 is usually chosen as X0 = (1, 1, , 1)T . But, if the
initial vector X0 is poor, then the formula (6.20) does not give 1 , i.e., the limit of the
(Y
)r
may not exist. If this situation occurs, then the initial vector must be
ratio (Xk+1
k )r
changed.

## Eigenvalues and Eigenvectors of a Matrix

377

Note 6.3.2 The power method is also used to nd the least eigenvalue of a matrix
A. If X is the eigenvector corresponding to the eigenvalue then AX = X. If A
is non-singular then A1 exist. Therefore, A1 (AX) = A1 X or, A1 X = 1 X.
This means that if is an eigenvalue of A then 1 is an eigenvalue of A1 and the same
eigenvector X corresponds to the eigenvalue 1/ of the matrix A1 . Thus, if is largest
(in magnitude) eigenvalue of A then 1/ is the least eigenvalue of A1 .
Note 6.3.3 We observed that the coecient Xj in (6.16) goes to zero in proposition
to (j /1 )k and that the speed of convergence is governed by the terms (2 /1 )k . Consequently, the rate of convergence is linear.
Example 6.3.2 Find the largest eigenvalue in magnitude and corresponding eigenvector of the matrix

1 3 2
A = 1 0 2
3 4 5
Solution. Let the initial vector be X0 = (1, 1, 1)T .
The rst iteration
is given
by

1 3 2
1
6
Y1 = AX0 = 1 0 2 1 = 1 .
3 4 5
1
12

0.50000
Y
1
= 0.08333 .
Therefore (1) = 12 and X1 =
12
1.0000

1 3 2
0.50000
2.75
Y2 = AX1 = 1 0 2 0.08333 = 1.5
3 4 5
1.00000
6.83333

0.40244
Y2
= 0.21951 .
(2) = 6.83333, X2 =
6.83333
1.0000

0.40244
3.06098
1 3 2
Y3 = AX2 = 1 0 2 0.21951 = 1.59756
3 4 5
1.00000
7.08537

0.43201
(3) = 7.08537, X3 = 0.22547 .
1.00000

3.10843
0.43185
Y4 = 1.56799 , X4 = 0.21784 , (4) = 7.19793.
7.19793
1.

## 378 Numerical Analysis

3.08691
0.43050
Y5 = 1.56950 , X5 = 0.21880 , (5) = 7.16691.
7.16672
1.

3.08691
0.43073
Y6 = 1.56950 , X6 = 0.21900 , (6) = 7.16672.
7.16672
1.

3.08772
0.43075
Y7 = 1.56927 , X7 = 0.21892 , (7) = 7.16818.
7.16818
1.0

3.08752
0.43074
Y8 = 1.56925 , X8 = 0.21893 , (8) = 7.16795.
7.16795
1.0

3.08752
0.43074
Y9 = 1.56926 , X9 = 0.21893 , (9) = 7.16792.
7.16792
1.0

3.08753
0.43074
Y10 = 1.56926 , X10 = 0.21893 , (10) = 7.16794.
7.16794
1.0
The required largest eigenvalue is 7.1679 correct up to four decimal places and the
corresponding eigenvector is

0.43074
0.21893 .
1.00000

Algorithm 6.2 (Power method). This method determines the largest eigenvalue
(in magnitude) and its corresponding eigenvector of a square matrix A.
Algorithm Power Method
Step 1. Read the matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. Set initial vector X0 = (1, 1, 1, . . . , 1)T of n components.
Step 3. Find the product Y = AX0 .
Step 4. Find the largest element (in magnitude) of the vector Y and let it be .
Step 5. Divide all the elements of Y by and take it as X1 , i.e., X1 = Y/.
Step 6. //Let X0 = (x01 , x02 , . . . , x0n ) and X1 = (x11 , x12 , . . . , x1n ).//
If |xoi x1i | > for at least i then
set X0 = X1 and goto Step 3.
Step 7. Print as largest eigenvalue and corresponding eigenvector X1 of A.
end Power Method

## Eigenvalues and Eigenvectors of a Matrix

Program 6.2
.
/* Program Power Method
This program finds the largest eigenvalue (in magnitude)
of a square matrix.*/
#include<stdio.h>
#include<math.h>
void main()
{
int n,i,j,flag;
float a,x0,x1,y,lambda,eps=1e-5;
printf("Enter the size of the matrix ");
scanf("%d",&n);
printf("Enter the elements row wise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("The given matrix is\n");
for(i=1;i<=n;i++){ /* printing of A */
for(j=1;j<=n;j++) printf("%f ",a[i][j]); printf("\n");
}
printf("\n");
for(i=1;i<=n;i++) {
x0[i]=1; x1[i]=1; /* initialization */
}
do
{
flag=0;
for(i=1;i<=n;i++) x0[i]=x1[i]; /* reset x0 */
for(i=1;i<=n;i++) /* product of A and X0 */
{
y[i]=0;
for(j=1;j<=n;j++) y[i]+=a[i][j]*x0[j];
}
lambda=y; /* finds maximum among y[i] */
for(i=2;i<=n;i++) if(lambda<y[i]) lambda=y[i];
for(i=1;i<=n;i++) x1[i]=y[i]/lambda;
for(i=1;i<=n;i++) if(fabs(x0[i]-x1[i])>eps) flag=1;
}while(flag==1);
printf("The largest eigenvalue is %8.5f \n",lambda);
printf("The corresponding eigenvector is \n");
for(i=1;i<=n;i++) printf("%8.5f ",x1[i]);
}/* main */

379

## 380 Numerical Analysis

A sample of input/output:
Enter the size of the matrix 5
Enter the elements row wise
3 4 5 6 7
0 0 2 1 3
3 4 5 -2 3
3 4 -2 3 4
0 1 2 0 0
The given matrix is
3.000000 4.000000 5.000000 6.000000 7.000000
0.000000 0.000000 2.000000 1.000000 3.000000
3.000000 4.000000 5.000000 -2.000000 3.000000
3.000000 4.000000 -2.000000 3.000000 4.000000
0.000000 1.000000 2.000000 0.000000 0.000000
The largest eigenvalue is 10.41317
The corresponding eigenvector is
1.00000 0.20028 0.62435 0.41939 0.13915
6.3.3

## It is mentioned earlier that if is the largest eigenvalue of A then 1/ is the smallest

eigenvalue of A and 1/ can be obtained by nding the largest eigenvalue of A1 . But,
computation of A1 is a labourious process, so a simple process is needed. One simple
method is introduced here.
If the largest eigenvalue (in magnitude) 1 , of an n n matrix A is known then the
smallest magnitude eigenvalue can be computed by using power method for the matrix

B = (A 1 I) instead of A. The eigenvalues of the matrix B are i = (i 1 ) (called

shifting eigenvalues), i = 1, 2, . . . , n, where i are the eigenvalues of A. Obviously, n is
the largest magnitude eigenvalue of B. Again, if Xn is the corresponding eigenvector,

then BXn = n Xn or, (A 1 I)Xn = (n 1 )Xn , i.e., AXn = n Xn . Hence Xn is
also the eigenvector of A, corresponding to the eigenvalue n .

6.4

## Eigenvalues for Symmetric Matrices

The methods discussed earlier are also applicable for symmetric matrices, but, due to
the special properties of symmetric matrices some ecient methods are developed here.
Among them three commonly used methods, viz., Jacobi, Givens and Householder are
discussed here.
In linear algebra it is established that all the eigenvalues of a real symmetric matrix
are real.

## Eigenvalues and Eigenvectors of a Matrix

6.4.1

381

Jacobis method

This method is widely used to nd the eigenvalues and eigenvectors of a real symmetric
matrix. Since all the eigenvalues of A are real and there exist a real orthogonal matrix
S such that S1 AS is a diagonal matrix D. As D and A are similar, the diagonal
elements of D are the eigenvalues of A. But, the computation of the matrix S is not a
simple task. It is obtained by a series of orthogonal transformations S1 , S2 , . . . , Sn , . . .
as discussed below.
Let |aij | be the largest element among the o-diagonal elements of A. Now, we
construct an orthogonal matrix S1 whose elements are dened as
sij = sin , sji = sin , sii = cos , sjj = cos ,

(6.21)

all other o-diagonal elements are zero and all other diagonal elements are unity. Thus
S1 is of the form

1
0

..
.

S1 = .
..

..
.
0

0
0

..
.

0 cos sin 0

..
..
..
..
.
.
.
.

0 sin cos 0

..
..
..
..
.
.
.
.
0
0

0
1
0
1
..
.

0
0
..
.

0
0
..
.

(6.22)

where cos , sin , sin and cos are at the positions (i, i), (i, j), (j, i) and (j, j) respectively.

aii aij
be a sub-matrix of A formed by the elements aii , aij , aji and ajj .
Let A1 =
aji ajj
an orthogonal transformation is applied which is
To reduce A1 to a diagonal matrix,

cos sin
, where is an unknown quantity and it will be selected
dened as S1 =
sin cos
in such a way that A1 becomes diagonal.
Now,
S1

=

=

A1 S1

cos sin
cos sin
aii aij
aji ajj
sin cos
sin cos

## aii cos2 + aij sin 2 + ajj sin2

(ajj aii ) sin cos + aij cos 2

## (ajj aii ) sin cos + aij cos 2

.
aii sin2 aij sin 2 + ajj cos2

## 382 Numerical Analysis

This matrix becomes a diagonal matrix if (ajj aii ) sin cos + aij cos 2 = 0,
That is , if
2aij
tan 2 =
.
(6.23)
aii ajj
The value of can be obtained from the following relation.


2aij
1
1
.
= tan
2
aii ajj

(6.24)

This expression gives four values of , but, to get smallest rotation, should lie in
/4 /4. The equation (6.24) is valid for all i, j if aii = ajj . If aii = ajj then

4 , if aij > 0
(6.25)
=

, if aij < 0.
4
1

Thus the o-diagonal elements sij and sji of S1 A1 S1 vanish and the diagonal
elements are modied. The rst diagonal matrix is obtained by computing D1 =
S1
1 A1 S1 . In the next step largest o-diagonal (in magnitude) element is selected
from the matrix D1 and the above process is repeated to generate another orthogonal
matrix S2 to compute D2 . That is,
1 1
1
A(S1 S2 ).
D2 = S1
2 D1 S2 = S2 (S1 AS1 )S2 = (S1 S2 )

## In this way, a series of two-dimensional rotations are performed. At the end of k

transformations the matrix Dk is obtained as
1
1
Dk = S1
k Sk1 S1 AS1 S2 Sk1 Sk

= (S1 S2 Sk )1 A(S1 S2 Sk )

= S1 AS

(6.26)

where S = S1 S2 Sk .
As k , Dk tends to a diagonal matrix. The diagonal elements of Dk are the
eigenvalues and the columns of S are the corresponding eigenvectors.
The method has a drawback. The elements those are transferred to zero during
diagonalisation may not necessarily remain zero during subsequent rotations. The value
of must be veried for its accuracy by checking whether | sin2 + cos2 1 | is
suciently small.
Note 6.4.1 It may be noted that for orthogonal matrix S, S1 = ST .

## Eigenvalues and Eigenvectors of a Matrix

383

Note 6.4.2 It can be shown that the minimum number of rotations required to transform a real symmetric matrix into a diagonal matrix is n(n 1)/2.
Example

1 2
A=2 1
2 2

6.4.1
Find the eigenvalues and eigenvectors of the symmetric matrix
2
2 using Jacobis method.
1

Solution. The largest o-diagonal element is 2 at (1, 2), (1, 3) and (2, 3) positions.

4
2a12
= = i.e., = .
The rotational angle is given by tan 2 =
a11 a22
0
4
Thus the orthogonal matrix
S
is
1

cos /4 sin /4 0
1/2 1/ 2 0
S1 = sin /4 cos /4 0 = 1/ 2 1/ 2 0 .
0
0
1
0
0
1
Then the rst rotation yields

1/ 2 1/2 0
1 2 2
1/2 1/ 2 0
1/ 2 1/ 2 0 2 1 2 1/ 2 1/ 2 0
D1 = S1
1 AS1 =
2 2 1
0
0 1
0
0
1

3
0 2.82843
3
0 4/ 2
.
0
1
0
= 0 1 0 =
2.82843
0
1
1
4/ 2 0
The largest o-diagonal element of D1 is now 2.82843 situated at (1, 3) position and
hence the rotational angle is
2a13
1
= 0.61548.
= tan1
2
a11 a33
The second orthogonal matrix S2 is

cos 0 sin
0.81650 0 0.57735
.
0 =
0
1
0
S2 = 0 1
sin 0 cos
0.57735 0 0.81650
Then second rotation gives
D1 S2
D2 = S1
2

0.81650 0 0.57735
3
0 2.82843
0.81650 0 0.57735

0
1
0
0
1
0
0
1
0
=
0.577351 0 0.81650
2.82843 0
1
0.57735 0 0.81650

5 0 0

= 0 1 0 .
0 0 1

## 384 Numerical Analysis

Thus D2 becomes a diagonal matrix and hence the eigenvalues are 5, 1, 1.
The eigenvectors are the columns of S, where

1/2 1/ 2 0
0.81650 0 0.57735

0
1
0
S = S1 S2 = 1/ 2 1/ 2 0
0.57735 0 0.81650
0
0
1

## 0.57735 0.70711 0.40825

= 0.57735 0.70711 0.40825 .
0.57735
0
0.81650
Hence the eigenvalues are 5, 1, 1 and the corresponding eigenvectors are
(0.57735, 0.57735, 0.57735)T , (0.70711, 0.70711, 0)T ,
(0.40825, 0.40825, 0.81650)T respectively.
Note that the eigenvectors are normalized and two independent eigenvectors (last two
vectors) for the eigenvalue 1 are obtained by this method.
In this problem, two rotations are used to convert A into a diagonal matrix. But,
this does not happen in general. The following example shows that at least six rotations
are needed to diagonalise a symmetric matrix.
Example

## 6.4.2 Find all the eigenvalues and eigenvectors of the matrix

2 3 1
3 2 2 by Jacobis method.
1 2 1

2 3 1
Solution. Let A = 3 2 2 .
1 2 1
It is a real symmetric matrix and the Jacobis method is applicable.
The largest o-diagonal element is at a12 = a21 and it is 3.

2a12
6
Then tan 2 =
= = , and this gives = .
a11 a22
0
4
Thus the orthogonal matrix S1 is

cos /4 sin /4 0
1/2 1/ 2 0
S1 = sin /4 cos /4 0 = 1/ 2 1/ 2 0 .
0
0
1
0
0
1
The rst rotation gives

1/ 2 1/2 0
2 3 1
1/2 1/ 2 0
1/ 2 1/ 2 0 3 2 2 1/ 2 1/ 2 0
= S1
1 AS1 =
1 2 1
0
0 1
0
0
1

D1

385

5
0 3/2

= 0
1
1/ 2 .
3/ 2 1/ 2 1

## The largest o-diagonal element of D1 is 3/ 2, situated at (1, 3) position. Then

6/ 2
2a13
= 1.06066. or, = 12 tan1 (0.69883) = 0.407413
=
tan 2 =
a11 a33
51

0.91815 0 0.39624
.
0
1
0
So, the next orthogonal matrix S2 is S2 =
0.39624 0 0.91815
D1 S2
D2 = S1
2

0.91815 0 0.39624
5
0
2.12132
0.91815 0 0.39624

0
1
0
0
1 0.70711
0
1
0
=
0.39624 0 0.91815
2.12132 0.70711
1
0.39624 0 0.91815

5.91548 0.28018
0
= 0.28018 1.0 0.64923 .
0
0.64923 0.08452
The largest o-diagonal element of D2 is 0.64923, present at the position (2, 3). Then
2a23
1
= 1.19727 or, = tan1 (1.19727) = 0.43747.
a22 a33
2

1
0
0
Therefore, S3 = 0 0.90583 0.42365 .
0 0.42365 0.90583

## 5.91548 0.25379 0.11870

0.25379 1.30364
.
0
D3 = S1
3 D2 S3 =
0.11870
0
0.38816
Again, largest o-diagonal element is 0.25379, located at (1, 2) position.
tan 2 =

2a12
1
tan1
= 0.03510.
2
a11 a22

0.99938 0.03509 0
S4 = 0.03509 0.99938 0 .
0
0
1

5.92439
0
0.11863

0
1.31255 0.00417 .
D4 = S1
4 D3 S4 =
0.11863 0.00417 0.38816

Therefore, =

## 386 Numerical Analysis

Here largest o-diagonal element is 0.11863 at (1, 3) position. In this case
2a13
1
= 0.02141.
= tan1
2
a11 a33

0.99977 0 0.02141
.
0
1
0
S5 =
0.02141 0 0.99977

5.92693 0.00009
0
0.00009 1.31255 0.00417 .
D5 = S1
5 D4 S5 =
0
0.00417 0.38562
The largest o-diagonal element in magnitude is 0.00417 situated at (2, 3) position.
Then
2a23
1
= 0.00246.
= tan1
2
a22 a33

1
0
0
1
0.00246 .
S6 = 0
0 0.00246
1

5.92693 0.00009
0
0.00009 1.31256
.
0
D6 = S1
6 D5 S6 =
0
0
0.38563
This matrix is almost diagonal and hence the eigenvalues are 5.9269, 1.3126 and
0.3856 correct up to four decimal places.
The eigenvectors are the columns of

## 0.61852 0.54567 0.56540

S = S1 S2 . . . S6 = 0.67629 0.73604 0.02948 .
0.40006 0.40061 0.82430
That is, the eigenvectors corresponding to the eigenvalues 5.9269, 1.3126 and 0.3856
are respectively (0.61825, 0.67629, 0.40006)T ,
(0.54567, 0.73604, 0.40061)T and (0.56540, 0.02948, 0.82430)T .
Note 6.4.3 This example shows that the elements which were annihilated by a rotation
may not remain zero during the next rotations.
Algorithm 6.3 (Jacobis method). This method determines the eigenvalues and
eigenvectors of a real symmetric matrix A, by converting A into a diagonal matrix
by similarity transformation.
Algorithm Jacobi
Step 1. Read the symmetric matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. Initialize D = A and S = I, a unit matrix.

## Eigenvalues and Eigenvectors of a Matrix

Step 3. Find the largest o-diagonal element (in magnitude) from D = [dij ] and
let it be dij .
Step 4. //Find the rotational angle .//
If dii = djj then
if dij > 0 then = /4 else = /4 endif;
else


2dij
1
1
;
= 2 tan
dii djj
endif;
Step 5. //Compute the matrix S1 = [spq ]//
Set spq = 0 for all p, q = 1, 2, . . . , n
skk = 1, k = 1, 2, . . . , n
and sii = sjj = cos , sij = sin , sji = sin .
Step 6. Find D = ST
1 D S1 and S = S S1 ;
Step 7. Repeat steps 3 to 6 until D becomes diagonal.
Step 8. Diagonal elements of D are the eigenvalues and the columns of S are
the corresponding eigenvectors.
end Jacobi
Program 6.3
.
/* Program Jacobis Method to find eigenvalues
This program finds all the eigenvalues and the corresponding
eigenvectors of a real symmetric matrix. Assume that the
given matrix is real symmetric. */
#include<stdio.h>
#include<math.h>
void main()
{
int n,i,j,p,q,flag;
float a,d,s,s1,s1t;
float temp,theta,zero=1e-4,max,pi=3.141592654;
printf("Enter the size of the matrix ");
scanf("%d",&n);
printf("Enter the elements row wise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("The given matrix is\n");
for(i=1;i<=n;i++) /* printing of A */
{
for(j=1;j<=n;j++) printf("%8.5f ",a[i][j]); printf("\n");
}
printf("\n");

387

## 388 Numerical Analysis

/* initialization of D and S */
for(i=1;i<=n;i++) for(j=1;j<=n;j++){
d[i][j]=a[i][j]; s[i][j]=0;
}
for(i=1;i<=n;i++) s[i][i]=1;
do
{
flag=0;
/* find largest off-diagonal element */
i=1; j=2; max=fabs(d);
for(p=1;p<=n;p++) for(q=1;q<=n;q++)
{ if(p!=q) /* off-diagonal element */
if(max<fabs(d[p][q])){
max=fabs(d[p][q]); i=p; j=q;
}
}
if(d[i][i]==d[j][j]){
if(d[i][j]>0) theta=pi/4; else theta=-pi/4;
}
else
{
theta=0.5*atan(2*d[i][j]/(d[i][i]-d[j][j]));
}
/* construction of the matrix S1 and S1T */
for(p=1;p<=n;p++) for(q=1;q<=n;q++)
{s1[p][q]=0; s1t[p][q]=0;}
for(p=1;p<=n;p++) {s1[p][p]=1; s1t[p][p]=1;}
s1[i][i]=cos(theta); s1[j][j]=s1[i][i];
s1[j][i]=sin(theta); s1[i][j]=-s1[j][i];
s1t[i][i]=s1[i][i];
s1t[j][j]=s1[j][j];
s1t[i][j]=s1[j][i];
s1t[j][i]=s1[i][j];
/* product of S1T and D */
for(i=1;i<=n;i++)
for(j=1;j<=n;j++){
temp[i][j]=0;
for(p=1;p<=n;p++) temp[i][j]+=s1t[i][p]*d[p][j];
}
/* product of temp and S1 i.e., D=S1T*D*S1 */
for(i=1;i<=n;i++)
for(j=1;j<=n;j++){

## Eigenvalues and Eigenvectors of a Matrix

d[i][j]=0;
for(p=1;p<=n;p++) d[i][j]+=temp[i][p]*s1[p][j];
}
/* product of S and S1 i.e., S=S*S1 */
for(i=1;i<=n;i++)
for(j=1;j<=n;j++)
{
temp[i][j]=0;
for(p=1;p<=n;p++) temp[i][j]+=s[i][p]*s1[p][j];
}
for(i=1;i<=n;i++) for(j=1;j<=n;j++) s[i][j]=temp[i][j];
for(i=1;i<=n;i++) for(j=1;j<=n;j++) /* is D diagonal ? */
{
if(i!=j) if(fabs(d[i][j]>zero)) flag=1;
}
}while(flag==1);
printf("The eigenvalues are\n");
for(i=1;i<=n;i++) printf("%8.5f ",d[i][i]);
printf("\nThe corresponding eigenvectors are \n");
for(j=1;j<=n;j++){
printf("(");
for(i=1;i<n;i++) printf("%8.5f,",s[i][j]);
printf("%8.5f)\n",s[n][j]);
}
}/* main */
A sample of input/output:
Enter the size of the matrix 4
Enter the elements row wise
1 2 3 4
2 -3 3 4
3 3 4 5
4 4 5 0
The given matrix is
1.000000 2.000000 3.000000 4.000000
2.000000 -3.000000 3.000000 4.000000
3.000000 3.000000 4.000000 5.000000
4.000000 4.000000 5.000000 0.000000
The eigenvalues are
-0.73369 -5.88321 11.78254 -3.16564

389

## The corresponding eigenvectors are

( 0.74263, 0.04635,-0.65234, 0.14421)
( 0.13467, 0.74460, 0.06235,-0.65081)
( 0.43846, 0.33395, 0.64097, 0.53422)
(-0.48797, 0.57611,-0.39965, 0.51987)
6.4.2

Let

0 0
0 0
b4 0
.. .. ..
. . .
0 0 0 0 bn

a1
b2

A= 0
..
.

b2
a2
b3
..
.

0
b3
a3
..
.

0
0
0
..
.

an

## be a symmetric tri-diagonal matrix. The characteristic equation of this matrix is

0
a1 b2
b2
a

b
2
3

0
a

b
3
3
pn () =

..
..
..

.
.
.
0
0
0

0 0
0
0 0
0

b4 0
0
= 0.

.. .. ..
..

. . .
.
0 bn an

Expanding by minors, the sequence {pn ()} satises the following equations.
p0 () = 1
p1 () = a1
pk+1 () = (ak+1 )pk () b2k+1 pk1 (), k = 1, 2, . . . , n

(6.27)

## The polynomial pn () is the characteristic polynomial of A.

If none of b2 , b3 , . . . , bn vanish, then {pn ()} is a Sturm sequence. Then using the
property of Sturm sequence, one can determine the intervals (containing the eigenvalue),
by substituting dierent values of . That is, if N () denotes the number of changes
in sign of the Sturm sequence for a given value of , then | N (a) N (b) | represents
the number of zeros (eigenvalues) lie in [a, b], provided pn (a) and pn (b) are not zero.
Once the location of an eigenvalue is identied then using any iterative method such as
bisection method, Newton-Raphson method etc. one can determine it.
Having computed the eigenvalues of A, the eigenvectors of A can be directly computed by solving the resulting homogeneous linear equations (A I)X = 0.

391

Example

## 6.4.3 Find the eigenvalues of the following tri-diagonal matrix

3 1 0
1 2 1 .
0 1 1
Solution. The Sturm sequence {pn ()} is given by
p0 ()
p1 ()
p2 ()
p3 ()

=
=
=
=

1
3
(2 )p1 () 1p0 () = 2 5 + 5
(1 )p2 () 1p1 () = 3 + 62 9 + 2.

1
0
1
2
3
4

p0
+
+
+
+
+
+

p1
+
+
+
+
0

p2
+
+
+

p3 N ()
+ 0
+ 0

1
0
1
+ 2

## Here p3 (2) = 0, so = 2 is an eigenvalue. The other two eigenvalues lie in the

intervals (0, 1) and (3, 4) as N (1) N (0) = 1 and N (4) N (3) = 1.
To nd the eigenvalue within (0, 1) using Newton-Raphson method
Let (i) be the ith approximate value of the eigenvalue in (0, 1). The Newton-Raphson
iteration scheme is
p3 ((i) )
(i+1) = (i)  (i) .
p3 ( )
Initially, let (0) = 0.5. The successive iterations are shown below.
(i)
p3 ((i) ) p3 ((i) ) (i+1)
0.5
1.12500 3.75000 0.20000
0.2
0.43200 6.72000 0.26429
0.26429 0.02205 6.03811 0.26794
0.26794 0.00007 6.00012 0.26795
0.26795 0.00000 6.00000 0.26795
Hence the other eigenvalue is 0.26795.
To nd the eigenvalue within (3, 4) using Newton-Raphson method
Let the initial eigenvalue be (0) = 3.5.

## 392 Numerical Analysis

The calculations are shown in the following table.
(i)
p3 ((i) ) p3 ((i) ) (i+1)
3.5
1.12500 3.75000 3.80000
3.80000 0.43200 6.72000 3.73572
3.73572 0.02206 6.03812 3.73206
3.73206 0.00007 6.00012 3.73205
3.73205 0.00000 6.00000 3.73205
The other eigenvalue is 3.73205.

Hence all the eigenvalues are 2, 0.26795 and 3.73205, the exact values are 2, 2 3.
6.4.3

Givens method

## The Givens method is used to nd eigenvalues of a real symmetric matrix A = [aij ].

This method preserves the zeros in the o-diagonal elements, once they are created. This
method works into two steps. In rst step, the given symmetric matrix is converted to
a symmetric tri-diagonal matrix using plane rotations. In second step, the eigenvalues
of this new matrix are determined by the method discussed in previous section.
The conversion to a tri-diagonal form is done by using orthogonal transformations as
in Jacobis method. In this case, the rotation is performed with the elements a22 , a23 , a32
and a33 .

1 0
0
0 0
0 cos sin 0 0

Let S1 =
0 sin cos 0 0 be the orthogonal matrix, where is unknown.
0 0
0
1 0
0 0
0
0 1
Let B1 be the transformed matrix under orthogonal transformation S1 , Then B1 =
S1
1 AS1 .
The elements of (1, 3) and (3, 1) positions are equal and they are a12 sin +a13 cos .
The angle is now obtained by putting this value to zero.
That is,
a13
or,
tan =
.
(6.28)
a12 sin + a13 cos = 0
a12
This transformation is now considered as a rotation in the (2,3)-plane. It may be
noted that this computation is more simple than Jacobis method. The matrix B1 has
the form


## a11 a12 0 a14 a1n

a21 a22 a23 a24 a2n





B1 =
0 a32 a33 a34 a3n .

an1 an2 an3 an4 ann

## Eigenvalues and Eigenvectors of a Matrix

393

Then apply the rotation in (2,4)-plane to convert a14 and a41 to zero. This would
not eect zeros that have been created earlier. Successive rotations are carried out in
the planes (2, 3), (2, 4), . . . , (2, n) where s are so chosen that the new elements at
the positions (1, 3), (1, 4), . . . , (1, n) vanish. After (n 2) such rotations, all elements
of rst row and column (except rst two) become zero. Then the transformed matrix
Bn2 after (n 2) rotations reduces to the following form:

Bn2

a11
a21

=
0
0
0

a12
a22
a32

an2

0
a23
a33

an3

0
a24
a34

an4

0
a2n
a3n

ann

The second row of Bn2 is taken in the same way as of the rst row. The rotations
are made in the planes (3,4), (3,5), . . . , (3, n). Thus, after (n 2) + (n 3) + + 1
(n 1)(n 2)
rotations the matrix A becomes a tri-diagonal matrix B of the form
=
2

a1 b2 0 0 0 0
b2 a2 b3 0 0 0

B=
0 b3 a3 b4 0 0 .

0 0 0 0 bn an
In this process, the previously created zeros are not eected by successive rotations.
The eigenvalues of B and A are same as they are similar matrices.
Example

3
A=
1

## 6.4.4Find the eigenvalues of the symmetric matrix

3 1
1 2 using Givens method.
2 1

## Solution. Let the orthogonal matrix S1 be

1 0
0
S1 = 0 cos sin ,
0 sin cos
1
a13
= , i.e., = 0.32175.
a12
3

1
0
0
Therefore, S1 = 0 0.94868 0.31623 .
0 0.31623 0.94868
where tan =

## 394 Numerical Analysis

The reduced matrix is

1
0
0
2 3 1
1
0
0
0 0.94868 0.31623 3 1 2 0 0.94868 0.31623
B = S1
1 AS1 =
0 0.31623 0.94868
1 2 1
0 0.31623 0.94868

2
3.16228 0.00001
2.2 .
= 3.16228 0.40001
0.00001
2.2
0.40001

2
3.1623 0
Let B = 3.1623 0.4 2.2 .
0
2.2 0.4
The Sturm sequence is
p0 ()
p1 ()
p2 ()
p3 ()

=
=
=
=

1
2
(0.4 )p1 () 3.16232 p0 () = 2 1.6 10.8
(0.4 )p2 () 2.22 p1 () = 3 + 22 + 15 14.

## Now, we tabulate the values of p0 , p1 , p2 , p3 for dierent values of .

4
3
1
0
1
2
4
5

p0
+
+
+
+
+
+
+
+

p1
+
+
+
+
+
0

p2
+
+

p3 N ()
+ 0

1
+ 2
+ 2
+ 2

From this table, it is observed that the eigenvalues are located in the intervals
(4, 3), (0, 1) and (4, 5). Any iterative method may be used to nd them. Using
Newton-Raphson method, we nd the eigenvalues 3.47531, 0.87584 and 4.59947 of
B. These are also the eigenvalues of A.

6.4.4

Householders method

This method is applicable to a real symmetric matrix of order nn. It is more economic
and ecient than the Givens method. Here also a sequence of orthogonal (similarity)
transformations is used on A to get a tri-diagonal matrix. Each transformation produces
a complete row of zeros in appropriate positions, without aecting the previous rows.

395

## Thus, (n 2) Householder transformations are needed to produce the tri-diagonal form.

The orthogonal transformation used in this method is of the form
S = I 2VVT

(6.29)

## where V = (s1 , s2 , . . . , sn )T is a column vector containing n components, such that

VT V = s21 + s22 + + s2n = 1.

(6.30)

## The matrix S is symmetric and orthogonal, since

ST = (I 2VVT )T = I 2VVT = S
and

## ST S = (I 2VVT )(I 2VVT )

= I 4VVT + 4VVT VVT
= I 4VVT + 4VVT = I.

Thus

[using (6.30)].

S1 AS = ST AS = SAS,

(6.31)
(6.32)

## since S is orthogonal and symmetric.

Let A1 = A and form a sequence of transformations
Ar = Sr Ar1 Sr , r = 2, 3, . . . , n 1,

(6.33)

## where Sr = I 2Vr VrT , Vr = (0, 0, . . . , 0, sr , sr+1 , . . . , sn )T and s2r + s2r+1 + + s2n = 1.

At the rst transformation, we nd sr s in such a way that the elements in the positions (1, 3), (1, 4), . . . , (1, n) of A2 become zero. Also, the elements in the corresponding
positions in the rst column becomes zero. Therefore, one rotation creates n 2 zeros
in the rst row and rst column. In the second rotation, the elements in the positions
(2, 4), (2, 5), . . . , (2, n) and (4, 2), (5, 2), . . . , (n, 2) reduce to zeros.
Thus (n 2) Householder transformations are required to obtain the tri-diagonal
matrix An1 . This method is illustrated using a 4 4 matrix A = [aij ]44 .
In the rst transformation, let V2 = (0, s2 , s3 , s4 )T such that
s22 + s23 + s24 = 1.

(6.34)

Now,

S2 = I 2VVT

1
0
0
0
0 1 2s22 2s2 s3 2s2 s4

=
0 2s2 s3 1 2s23 2s3 s4 .
0 2s2 s4 2s3 s4 1 2s24

(6.35)

## 396 Numerical Analysis

The rst rows of A1 and S2 A1 are same. The elements in the rst row of A2 =
S2 A1 S2 are given by
a11 = a11
a12 = (1 2s22 )a12 2s2 s3 a13 2s2 s4 a14 = a12 2s2 p1
a13 = 2s2 s3 a12 + (1 2s23 )a13 2s3 s4 a14 = a13 2s3 p1
and a14 = 2s2 s4 a12 2s3 s4 a13 + (1 2s24 )a14 = a14 2s4 p1
where p1 = s2 a12 + s3 a13 + s4 a14 .
It can be veried that
a 11 + a 12 + a 13 + a 14 = a211 + a212 + a213 + a214 .
2

That is,
a 12 + a 13 + a 14 = a212 + a213 + a214 = q 2 (say)
2

(6.36)

## and q is a known quantity.

Since the elements at the positions (1, 3) and (1, 4) of A2 need to be zeros, a13 =
0, a14 = 0.
Hence

and

a13 2p1 s3 = 0

(6.37)

a14 2p1 s4 = 0

(6.38)

a12

(6.39)

## Multiplying equations (6.39), (6.37) and (6.38) by s2 , s3 and s4 respectively and

adding them to obtain the equation
p1 2p1 (s22 + s23 + s24 ) = qs2 , or, p1 = s2 q.

(6.40)

Thus from (6.39), (6.37) and (6.38) the values of s2 , s3 and s4 are obtained as


1
a13
a14
a12
2
,
s4 =
.
(6.41)
s2 =
1
,
s3 =
2
q
2s2 q
2s2 q
It is noticed that the values of s3 and s4 depend on s2 , so the better accuracy can
be achieved if s2 becomes large. This can be obtained by taking suitable sign in (6.41).
Choosing


a12 sign(a12 )
1
2
1+
.
(6.42)
s2 =
2
q

## Eigenvalues and Eigenvectors of a Matrix

397

The sign of the square root is irrelevant and positive sign is taken. Hence
s3 =

a13 sign(a12 )
,
2q s2

s4 =

a14 sign(a12 )
.
2q s2

Thus rst transformation generates two zeros in the rst row and rst column. The
second transformation is required to create zeros at the positions (2, 4) and (4, 2).
In the second transformation, let V3 = (0, 0, s3 , s4 )T and the matrix

1 0
0
0
0 1

0
0

(6.43)
S3 =
0 0 1 2s23 2s3 s4 .
0 0 2s3 s4 1 2s24
The values of s3 and s4 are to be computed using the previous technique. The new
matrix A3 = S3 A2 S3 is obtained. The zeros in rst row and rst column remain
unchanged while computing A3 . Thus, A3 becomes to a tri-diagonal form in this case.
The application of this method for a general n n matrix is obvious. The elements of
the vector Vk = (0, , 0, sk , sk+1 , , sn ) at the kth transformation are given by
. 


. n
akr sign(akr )
1
2
1+
,
r = k + 1, where q = /
a2ki
sk =
2
q
i=k+1

si =

aki sign(akr )
,
2qsk

i = k + 1, . . . , n.

Since the tri-diagonal matrix An1 is similar to the original matrix A, they have
identical eigenvalues. The eigenvalues of An1 are computed in the same way as in the
Givens method. Once the eigenvalues become available, the eigenvectors are obtained
by solving the homogeneous system of equations (A I)X = 0.
Example
6.4.5 Use the

## Householder method to reduce the matrix

2 1 1 1
1 4 1 1

A=
1 1 3 1 into the tri-diagonal form.
1 1 1 2
Solution. First rotation.




(1)(1)
1
2

= 0.78868,
s2 = 0.88807,
1+
s2 =
2
3
(1)(1)
= 0.32506,
s3 =
2 3 0.88807

1 (1)
s4 =
= 0.32506.
2 3 0.88807

## 398 Numerical Analysis

V2 = (0, 0.88807, 0.32506, 0.32506)T .
S2 = I 2V2 V2T

1
0
0
0
0 0.57734 0.57735 0.57735

=
0 0.57735 0.78867 0.21133 .
0 0.57735 0.21133 0.78867

2
1.73204
0
0
1.73204
5.0
0.21132 0.78867
.
A2 = S2 A1 S2 =

0
0.21132 2.28867
0.5
0
0.78867 0.5
1.71133
Second transformation. 
V3 = (0, 0, s3 , s4 )T . q = a223 + a224 = 0.81649,


a23 sign(a23 )
1
2
1+
= 0.62941,
s3 = 0.79335,
s3 =
2
q
a24 sign(a23 )
= 0.60876.
2qs3
V3 = (0, 0, 0.79335, 0.60876)T .

s4 =

S3

A3

1 0
0
0
0 1

0
0

= I 2V3 V3T =
0 0 0.25881 0.96592 .
0 0 0.96592 0.25882

2
1.73204
0
0
1.73204

5.0
0.81648
0
.
= S3 A2 S3 =

0
0.81648
2
0.57731
0
0
0.57731
2

## This is the required tri-diagonal matrix similar to A.

Algorithm 6.4 (Householder method). This method converts a real symmetric
matrix A of order n n into a real symmetric tri-diagonal form.
Algorithm Householder
Step 1. Read the symmetric matrix A = [aij ], i, j = 1, 2, . . . , n.
Step 2. Set k = 1, r = 2.
T
Step 3. //Compute the vector V = (v+
1 , v2 , , vn ) //
n
2
Step 3.1.
Compute q =
i=k+1 aki .
Step 3.2.
Set vi = 0 for i = 1, 2, . . . , r 1.

## Eigenvalues and Eigenvectors of a Matrix

Step 3.3.

Compute vr2 =

akr sign(akr )
1
2 1+
q
aki sign(akr )
2qvr


.

Compute vi =
for i = r + 1, . . . , n.
Step 4. Compute the transformation matrix S = I 2V VT .
Step 5. Compute A = S A S.
Step 6. Set k = k + 1, r = r + 1.
Step 7. Repeat steps 3 to 6 until k n 2.
end Householder
Step 3.4.

Program 6.4
.
/* Program Householder method
This program reduces the given real symmetric matrix
into a real symmetric tri-diagonal matrix. Assume that
the given matrix is real symmetric. */
#include<stdio.h>
#include<math.h>
void main()
{
int n,i,j,r=2,k,l,sign;
float a,v,s,temp,q;
printf("Enter the size of the matrix ");
scanf("%d",&n);
printf("Enter the elements row wise ");
for(i=1;i<=n;i++) for(j=1;j<=n;j++) scanf("%f",&a[i][j]);
printf("The given matrix is\n");
for(i=1;i<=n;i++) /* printing of A */
{
for(j=1;j<=n;j++) printf("%8.5f ",a[i][j]); printf("\n");
}
for(k=1;k<=n-2;k++)
{
q=0;
for(i=k+1;i<=n;i++) q+=a[k][i]*a[k][i];
q=sqrt(q);
for(i=1;i<=r-1;i++) v[i]=0;
sign=1; if(a[k][r]<0) sign=-1;
v[r]=sqrt(0.5*(1+a[k][r]*sign/q));
for(i=r+1;i<=n;i++) v[i]=a[k][i]*sign/(2*q*v[r]);
/* construction of S */
for(i=1;i<=n;i++) for(j=1;j<=n;j++) s[i][j]=-2*v[i]*v[j];

399

## 400 Numerical Analysis

for(i=1;i<=n;i++) s[i][i]=1+s[i][i];
for(i=1;i<=n;i++) for(j=1;j<=n;j++)
{
temp[i][j]=0;
for(l=1;l<=n;l++) temp[i][j]+=s[i][l]*a[l][j];
}
for(i=1;i<=n;i++) for(j=1;j<=n;j++)
{
a[i][j]=0;
for(l=1;l<=n;l++) a[i][j]+=temp[i][l]*s[l][j];
}
r++;
} /* end of loop k */
printf("The reduced symmetric tri-diagonal matrix is\n");
for(i=1;i<=n;i++)
{
for(j=1;j<=n;j++) printf("%8.5f ",a[i][j]);
printf("\n");
}
}/* main */
A sample of input/output:
Enter the size of the matrix 5
Enter the elements row wise
1 -1 -2 1 1
-1 0 1 3 2
-2 1 3 1 1
1 3 1 4 0
1 2 1 0 5
The given matrix is
1.00000 -1.00000 -2.00000 1.00000
-1.00000 0.00000 1.00000 3.00000
-2.00000 1.00000 3.00000 1.00000
1.00000 3.00000 1.00000 4.00000
1.00000 2.00000 1.00000 0.00000

1.00000
2.00000
1.00000
0.00000
5.00000

## The reduced symmetric tri-diagonal matrix is

1.00000 2.64575 -0.00000 0.00000 -0.00000
2.64575 1.00000 2.03540 -0.00000 0.00000

-0.00000
0.00000
-0.00000

6.5

401

## 2.03540 -0.58621 0.94489 0.00000

0.00000 0.94489 5.44237 -1.26864
0.00000 0.00000 -1.26864 6.14384

Exercise

## 1. Compare Jacobi, Givens and Householder methods to nd the eigenvalues of a

real symmetric matrix.
2. If X is any vector and P = I 2XXT , show that P is symmetric. What additional
condition is necessary in order that P is orthogonal ?
3. Use the Leverrier-Faddeev method to nd the characteristic equations of the following matrices.

2 1 3 4
1 1 1 1
2 5 7
3 2 4 1
1 1 1 1

(a) 6 3 4 , (b)
5 3 2 2 , (c) 1 1 1 1 .
5 2 3
3 3 1 1
1 1 1 1
4. Use the Leverrier-Faddeev method to nd the eigenvalues and eigenvectors of the
matrices

1 2 1 2

5 6 3
1 2 3
2 1 2 1
2 3

## , (b) 1 0 1 , (c) 3 1 2 , (d)

(a)
1 1 2 2 .
1 2
1 2 1
1 0 1
2 2 1 1
5. Find all eigenvalues
matrices using Rutishauser method.
of the following

3 2 3
4 5
(a)
, (b) 0 1 1 .
1 1
1 2 1
6. Use power method to nd the largest and the least (in magnitude) eigenvalues of
the following matrices.

4 1 0
2 1 2
3 1 0
1 2
(a)
, (b) 1 2 1 , (c) 5 3 3 , (d) 1 2 2 .
2 3
0 1 1
1 0 2
0 1 1
7. Use Jacobis method to nd the eigenvalues
of the

following matrices.

4 3 2 1
3 2 1
2 2 6
3 4 3 2

2 3 4 3 .
1 2 3
6 4 1
1 2 3 4

## 402 Numerical Analysis

8. UseGivens method

to
3 2 1
2

(a) 2 3 2 , (b) 2
1 2 3
6

nd
the eigenvalues
of the

## following symmetric matrices.

2 6
1 1 1
5 4 , (c) 1 2 3 .
4 1
1 3 1

## 9. Use Householder method to convert the above matrices to tri-diagonal form.

10. Find the eigenvalues
following matrices using Householder method.

of the
4 3 2
5 4 3
(a) 3 4 3 , (b) 4 5 4 .
2 3 4
3 4 5
11. Find the eigenvalues of the following
tri-diagonal
matrices.

4 1 0 0
2 1 0
3 1 0
1 4 1 0

## (a) 1 3 1 , (b) 1 3 1 , (c)

0 1 4 1 .
0 1 2
0 1 3
0 0 1 4
the following matrices.

2 1 1
3 7 2
(a) 6 1 3 , (b) 12 20 6 .
12 2 8
20 31 9

1 2 2 2
2 2 1 2

## 13. Transform the symmetric matrix A =

2 1 2 2 to a tri-diagonal

2 2 3
2
form, using Givens method, by a sequence of orthogonal transformations. Use
exact arithmetic.

Chapter 7

7.1

Dierentiation

## Numerical dierentiation is a method to nd the derivatives of a function at some values

of independent variable x, when the function f (x) is not known explicitly, but is known
only for a set of arguments.
Like interpolation, a number of formulae for dierentiation are derived. The choice
of formula depends on the point at which the derivative is to be determined. So, to nd
the derivative at a point at the beginning of the table, the formula based on Newtons
forward interpolation is used, but, at a point which is at the end of the table, the
formula based on Newtons backward interpolation is used. If the given values of xi are
not equispaced then the formula based on Lagranges interpolation is appropriate.
7.1.1

## The error in polynomial interpolation is

E(x) = (x x0 )(x x1 ) (x xn )

f n+1 ()
f n+1 ()
= w(x)
(n + 1)!
(n + 1)!

## where min{x, x0 , . . . , xn } < < max{x, x0 , . . . , xn } and w(x) = (x x0 )(x x1 ) (x

xn ). Obviously, = (x) is an unknown function of x.
Therefore,
E  (x) = w (x)

f n+2 () 
f n+1 ()
+ w(x)
(x).
(n + 1)!
(n + 1)!

(7.1)

The bound of the second term is unknown due to the presence of the unknown
quantity  (x).
403

## 404 Numerical Analysis

But, at x = xi , w(x) = 0. Thus,
E  (xi ) = w (xi )

f n+1 (i )
,
(n + 1)!

(7.2)

where min{x, x0 , . . . , xn } < i < max{x, x0 , . . . , xn }. The error can also be expressed
in terms of divided dierence.
f n+1 ()
.
Let E(x) = w(x)f [x, x0 , x1 , . . . , xn ] where f [x, x0 , x1 , . . . , xn ] =
(n + 1)!


Then E (x) = w (x)f [x, x0 , x1 , . . . , xn ] + w(x)f [x, x, x0 , x1 , . . . , xn ].
Now, this expression is dierentiated (k 1) times by Leibnitzs theorem.
k

E (x) =

k


Cr w(i) (x)

  
Cr w (x)(k i)!f [x, x, . . . , x, x0 , . . . , xn ]

i=0

k


dk1
(f [x, x0 , . . . , xn ])
dxki

ki+1

(i)

i=0

k

k!
i=0

i!

ki+1

  
w(i) (x)(k i)!f [x, x, . . . , x, x0 , . . . , xn ],

(7.3)

## where w(i) (x) denotes the ith derivative of w(x).

Note 7.1.1 If a function f (x) is well-approximated by a polynomial (x) of degree at
most n, the slope f  (x) can also be approximated by the slope  (x). But, the error
committed in  (x) is more than the error committed in (x).

7.2

Polynomial

## Suppose the function y = f (x) is known at (n + 1) equispaced points x0 , x1 , . . . , xn and

they are y0 , y1 , . . . , yn respectively, i.e., yi = f (xi ), i = 0, 1, . . . , n. Let xi = x0 + ih and
x x0
u=
, h is the spacing.
h
The Newtons forward interpolation formula is
u(u 1) 2
u(u 1) (u n 1) n
y0 + +
y0
2!
n!
u2 u 2
u3 3u2 +2u 3
u4 6u3 +11u2 6u 4
y0 +
y0 +
y0
= y0 +uy0 +
2!
3!
4!
u5 10u4 + 35u3 50u2 + 24u 5
y0 +
+
(7.4)
5!

(x) = y0 + uy0 +

with error
E(x) =

405

h
f
(),
(n + 1)!

## where min{x, x0 , , xn } < < max{x, x0 , . . . , xn }.

Dierentiating (7.4) successively with respect to x, we obtain

1
2u 1 2
3u2 6u + 2 3
4u3 18u2 +22u 6 4

y0 +
y0 +
y0 +
y0
(x) =
h
2!
3!
4!

y0 +
+
(7.5)
5!


du
1
as
=
dx
h

1
6u 6 3
12u2 36u + 22 4

y0 +
y0
(x) = 2 2 y0 +
h
3!
4!

## 20u3 120u2 + 210u 100 5

y0 +
+
(7.6)
5!

1
24u 36 4
60u2 240u + 210 5

3
y0 +
y0 +
(7.7)
(x) = 3 y0 +
h
4!
5!
and so on.
It may be noted that y0 , 2 y0 , 3 y0 , are constants.
The above equations give the approximate derivative of f (x) at arbitrary point x (=
x0 + uh).
When x = x0 , u = 0, the above formulae become

1
1
1
1
1
(7.8)
 (x0 ) =
y0 2 y0 + 3 y0 4 y0 + 5 y0
h
2
3
4
5

1
11
5
(7.9)
 (x0 ) = 2 2 y0 3 y0 + 4 y0 5 y0 +
h
12
6


1
3
7
(7.10)
 (x0 ) = 3 3 y0 4 y0 + 5 y0
h
2
4
and so on.
Error in dierentiation formula based on Newtons forward interpolation
polynomial
The error in Newtons forward interpolation formula is
E(x) = u(u 1) (u n)hn+1

f n+1 ()
.
(n + 1)!

## 406 Numerical Analysis

Then
 1 u(u 1) (u n) n+1 d n+1
f n+1 () d 
u(u 1) (u n) +
h
[f
()]
(n + 1)! du
h
(n + 1)!
dx
u(u 1) (u n) n+1 n+2
f n+1 () d
[u(u 1) (u n)] +
h
= hn
f
(1 ), (7.11)
(n + 1)! du
(n + 1)!

E  (x) = hn+1

## where min{x, x0 , x1 , . . . , xn } < , 1 < max{x, x0 , x1 , . . . , xn }.

Error at the point x = x0 , i.e., u = 0 is
f n+1 () d
hn (1)n n! f n+1 ()
[u(u 1) (u n)]u=0 =
(n + 1)! du
(n + 1)!
d
[u(u 1) (u n)]u=0 = (1)n n!]
[as
du
(1)n hn f n+1 ()
,
=
n+1

E  (x0 ) = hn

(7.12)

## where min{x, x0 , x1 , . . . , xn } < < max{x, x0 , x1 , . . . , xn }.

d2 y
dy
and 2 at the point
Example 7.2.1 From the following table nd the value of
dx
dx
x = 1.5.
x : 1.5
2.0
2.5
3.0
3.5
4.0
y : 3.375 7.000 13.625 24.000 38.875 59.000
Solution. The forward dierence table is
x
y
1.5 3.375

2 y

3 y

3.625
2.0 7.000

3.000
6.625

2.5 13.625

0.750
3.750

10.375
3.0 24.000

0.750
4.500

14.875
3.5 38.875

5.250
20.125

4.0 59.000

0.750

407

## Here x0 = 1.5 and h = 0.5. Then u = 0 and hence


1
1
1
y  (1.5) =
y0 2 y0 + 3 y0
h
2
3

1 
1
1
=
3.625 3.000 + 0.750 = 4.750.
0.5
2
3
1
1
(3.000 0.750) = 9.000.
y  (1.5) = 2 (2 y0 3 y0 + ) =
h
(0.5)2

7.3

Polynomial

## Suppose the function y = f (x) is known at (n + 1) points x0 , x1 , . . . , xn , i.e., yi = f (xi ),

x xn
i = 0, 1, 2, . . . , n are known. Let xi = x0 + ih, i = 0, 1, 2, . . . , n and v =
. Then
h
Newtons backward interpolation formula is
v(v + 1) 2
v(v + 1)(v + 2) 3
yn +
yn
2!
3!
v(v + 1)(v + 2)(v + 3) 4
v(v+1)(v+2)(v+3)(v + 4) 5
yn +
yn +
+
4!
5!

(x) = yn + vyn +

## The above equation is dierentiated with respect to x successively.

1
2v + 1 2
3v 2 + 6v + 2 3
4v 3 +18v 2 +22v+6 4
yn +
yn +
yn +
yn
h
2!
3!
4!

5v 4 + 40v 3 + 105v 2 + 100v + 24 5
yn +
+
(7.13)
5!
1
6v + 6 3
12v 2 + 36v + 22 4
yn +
yn
 (x) = 2 2 yn +
h
3!
4!

20v 3 + 120v 2 + 210v + 100 5
yn +
+
(7.14)
5!

1
24v + 36 4
60v 3 + 240v + 210 5
yn +
yn +
 (x) = 3 3 yn +
(7.15)
h
4!
5!
 (x) =

and so on.
The above formulae are used to determine the approximate dierentiation of rst,
second and third, etc. order at any point x where x = xn + vh.

## 408 Numerical Analysis

If x = xn then v = 0. In this case, the above formulae become

1
1
1
1
1
 (xn ) =
yn + 2 yn + 3 yn + 4 yn + 5 yn +
h
2
3
4
5

1 2
11 4
5 5

3
(xn ) = 2 yn + yn + yn + yn +
h
12
6


1
3
7
 (xn ) = 3 3 yn + 4 yn + 5 yn +
h
2
4

(7.16)
(7.17)
(7.18)

## Error in dierentiation formula based on Newtons backward interpolation

polynomial
The error in Newtons backward interpolation formula is
E(x) = v(v + 1)(v + 2) (v + n)hn+1
where v =
Then

f n+1 ()
,
(n + 1)!

x xn
and min{x, x0 , x1 , . . . , xn } < < max{x, x0 , x1 , . . . , xn }.
h
f n+1 ()
d
[v(v + 1)(v + 2) (v + n)]
dv
(n + 1)!
v(v + 1)(v + 2) (v + n) n+2
(1 ),
f
+hn+1
(n + 1)!

E  (x) = hn

## where min{x, x0 , x1 , . . . , xn } < , 1 < max{x, x0 , x1 , . . . , xn }.

Error at x = xn , i.e., at v = 0 is
f n+1 ()
d
[v(v + 1)(v + 2) (v + n)]
dv
(n + 1)!


d
n!
f n+1 ()
[v(v + 1) (v + n)]v=0 = n!
as
= hn
(n + 1)!
dv
n
n+1
h f
()
=
.
n+1

E  (xn ) = hn

(7.19)

## Example 7.3.1 A particle is moving along a straight line. The displacement x at

some time instances t are given below:
t : 0 1 2 3 4
x : 5 8 12 17 26
Find the velocity and acceleration of the particle at t = 4.

409

## Solution. The backward dierence table is

t
0
1
2
3
4

x
5
8
12
17
26

x 2 x 3 x 4 x
3
4
5
9

1
1
4

0
3

The velocity is

1
1
1
1
dx
=
xn + 2 xn + 3 xn + 4 xn +
dt
h
2
3
4

1
1
1
1
= 9+ 4+ 3+ 3
1
2
3
4
= 12.75.
The acceleration is

d2 x
1 2
11 4
3

=
x
+

x
+
x
+

n
n
n
dt2
h2
12

1
11
= 2 4+3+
3 = 9.75.
1
12

Example 7.3.2 A slider in a machine moves along a xed straight rod. Its distance
x cm along the rod are given in the following table for various values of the time t
(in second).
t (sec) : 1.0
1.1
1.2
1.3
1.4
1.5
x (cm) : 16.40 19.01 21.96 25.29 29.03 33.21
Find the velocity and the acceleration of the slider at time t = 1.5.
Solution. The backward dierence table is
t
1.0
1.1
1.2
1.3
1.4
1.5
Here h = 0.1.

x
16.40
19.01
21.96
25.29
29.03
33.21

x 2 x 3 x 4 x 5 x
2.61
2.95
3.33
3.74
4.18

0.34
0.38 0.04
0.41 0.03 0.01
0.44 0.03 0.00 0.01

## 410 Numerical Analysis

1
1 2 1 3 1 4 1 5
dx
=
+ + + + + xn
dt
h
2
3
4
5

1
1
1
1
1
4.18 + 0.44 + 0.03 + 0.00 + 0.01
=
0.1
2
3
4
5
= 44.12.

1
11 5
d2 x
2
3
= 2 + + + xn
dt2
h
12

11
5
1
0.44
+
0.03
+

0.00
+

0.01
=
(0.1)2
12
6
= 47.83.
Hence velocity and acceleration are respectively 44.12 cm/sec and 47.83 cm/sec2 .

7.4

## Suppose yi = f (xi ), i = 0, 1, . . . , n are given for 2n + 1 equispaced points x0 , x1 ,

x2 , . . ., xn , where xi = x0 ih, i = 0, 1, . . . , n.
The Stirlings interpolation polynomial is
u y1 + y0 u2
u3 u 3 y2 + 3 y1

+
2 y1 +

1!
2
2!
3!
2
5 5u3 + 4u 5 y
5y
u4 u2
u
+

3
2
+
4 y2 +

+
4!
5!
2
x x0
.
where u =
h

(x) = y0 +

## This equation is dierentiated with respect to x successively.

1 y1 + y0
3u2 1 3 y2 + 3 y1

(x) =
+ u2 y1 +

h
2
6
2

3
4
2
5
5
2u u 4
5u 15u + 4 y3 + y2
+
y2 +

+ .
12
120
2

3
3
2
1
y2 + y1 6u 1
 (x) = 2 2 y1 + u
+
4 y2
h
2
12

2u3 3u 5 y3 + 5 y2

+ .
+
12
2

(7.20)

(7.21)

(7.22)

At x = x0 , u = 0. Then

1 5 y2 + 5 y3
1 y0 + y1 1 3 y1 + 3 y2


+
+ . (7.23)
(x0 ) =
h
2
6
2
30
2

## Differentiation and Integration

and

1
1 4
2
(x0 ) = 2 y1 y2 + .
h
12


411

(7.24)

## Error in dierentiation formula based on Stirlings interpolation polynomial

The error of Stirlings interpolation formula is
E(x) = u(u2 12 )(u2 22 ) (u2 n2 )h2n+1

f 2n+1 ()
,
(2n + 1)!

## where min{x, xn , . . . , x0 , . . . , xn } < < max{x, xn , . . . , x0 , . . . , xn }.

Then
d
du
f 2n+1 ()
[u(u2 12 )(u2 22 ) (u2 n2 )] h2n+1
du
dx
(2n + 1)!
2n+1

d f
()
+h2n+1 [u(u2 12 )(u2 22 ) (u2 n2 )]
dx (2n + 1)!
f 2n+1 ()
d
= h2n [u(u2 12 )(u2 22 ) (u2 n2 )]
du
(2n + 1)!
f 2n+2 (1 )
,
+h2n+1 [u(u2 12 )(u2 22 ) (u2 n2 )]
(2n + 1)!

E  (x) =

(7.25)

## where min{x, xn , . . . , x0 , . . . , xn } < , 1 < max{x, xn , . . . , x0 , . . . , xn }.

d
At x = x0 , u = 0. Then
[u(u2 12 )(u2 22 ) (u2 n2 )] = (1)n (n!)2 .
du
In this case,
E  (x0 ) =

## (1)n (n!)2 2n 2n+1

h f
().
(2n + 1)!

(7.26)

Example 7.4.1 Compute the values of (i) f  (3), (ii) f  (3), (iii) f  (3.1), (iv) f  (3.1)
using the following table.
x :
1
2
3
4
5
f (x) : 0.0000 1.3863 3.2958 5.5452 8.0472

## 412 Numerical Analysis

Solution. The central dierence table is
x
y = f (x)
x2 = 1 0.0000

2 y

3 y

4 y

1.3863
x1 = 2

1.3863

0.5232
1.9095

x0 = 3

3.2958

0.1833
0.3399

2.2494
x1 = 4

5.5452

x2 = 5

8.0472

0.0960
0.0873

0.2526
2.5020

Since x = 3 and x = 3.1 are the middle of the table, so the formula based on central
dierence may be used. Here Stirlings formula is used to nd the derivatives.
(i) Here x0 = 3, h = 1, u = 0.
Then

1 y1 + y0 3 y2 + 3 y1


+
f (3) =
h
2
12

## 1 1.9095 + 2.2494 0.1833 0.0873

= 2.1020.
=
1
2
12

1
1
1 4
1

2
(ii) f (3) = 2 y1 y2 + = 2 0.3399
0.0960 = 0.3319.
h
12
1
12
= 0.1. Then
(iii) Let x0 = 3, h = 1, u = 3.13
1

1 y1 + y0
3u2 1 3 y2 + 3 y1
f (3.1) =
+ u2 y1 +
h
2
6
2

3
2u u 4
+
y2 +
12

3 (0.1)2 1 0.1833 0.0873
1 1.9095 + 2.2494
+ 0.1 0.3399 +
=
1
2
6
2

3
2 (0.1) 0.1
0.0960 = 2.1345.
+
12


(iv)

1
3 y2 + 3 y1 6u2 1 4
2
f (3.1) = 2 y1 + u
+
y2 +
h
2
12

0.1833 0.0873 6(0.1)2 1
1
+
0.0960
= 2 0.3399 + 0.1
1
2
12
= 0.31885.


7.5

413

## The Lagranges interpolation formula is

(x) = w(x)

n

i=0

yi
, where w(x) = (x x0 )(x x1 ) (x xn ).
(x xi )w (xi )

Then
 (x) = w (x)

n

i=0

## and  (x) = w (x)


yi
yi

w(x)
.

(x xi )w (xi )
(x xi )2 w (xi )
n

(7.27)

i=0

n



yi
yi


2w
(x)

(x xi )w (xi )
(x xi )2 w (xi )
n

i=0
n


+2w(x)

i=0

i=0

yi
.
(x xi )3 w (xi )

(7.28)

## The value of w (x) can be determined as

n


(x x0 )(x x1 ) (x xj1 )(x xj+1 ) (x xn ).
w (x) =
j=0

The formulae (7.27) and (7.28) are valid for all x except x = xi , i = 0, 1, . . . , n.
To nd the derivatives at the points x0 , x1 , . . . , xn , the Lagranges polynomial is
rearranged as
(x) = w(x)

n

i=0
i=j

yi
(x xi )w (xi )

## (x x0 )(x x1 ) (x xj1 )(x xj+1 ) (x xn )

yj .
(xj x0 )(xj x1 ) (xj xj1 )(xj xj+1 ) (xj xn )

Therefore,
 (xj ) = w (xj )

n

i=0
i=j

 yj
yi
,
+
(xj xi )w (xi )
xj xi
i=0
n

i=j

n

(xj xi ).
=
i=0
i=j

(7.29)

Note that

## (x x0 )(x x1 ) (x xj1 )(x xj+1 ) (x xn )

d
dx (xj x0 )(xj x1 ) (xj xj1 )(xj xj+1 ) (xj xn ) x=xj
=

n

i=0
i=j

1
.
xj xi

## This formula is used to nd the derivative at the points x = x0 , x1 , . . . , xn .

Error in dierentiation formula based on Lagranges interpolation
polynomial
The error term is
E(x) = w(x)

f n+1 ()
, where w(x) = (x x0 )(x x1 ) (x xn ).
(n + 1)!

Thus
E  (x) = w (x)

f n+2 (1 )
f n+1 ()
+ w(x)
,
(n + 1)!
(n + 1)!

(7.30)

## where min{x, x0 , x1 , . . . , xn } < , 1 < max{x, x0 , x1 , . . . , xn }.

At x = xi , i = 0, 1, 2, . . . , n,
E  (xi ) = w (xi )

f n+1 ()
.
(n + 1)!

(7.31)

## Example 7.5.1 Use the dierentiation formula based on Lagranges interpolation

to nd the value of f  (2) and f  (2.5) from the following table.
x : 2 3 5
6
y : 13 34 136 229
Solution. Here x0 = 2, x1 = 3, x2 = 5, x3 = 6.
w(x) = (x x0 )(x x1 )(x x2 )(x x3 ) = (x 2)(x 3)(x 5)(x 6).
w (x) = (x 3)(x 5)(x 6) + (x 2)(x 5)(x 6) + (x 2)(x 3)(x 6)
+ (x 2)(x 3)(x 5).
By the formula (7.29),
3
3


yi
y0
+
f  (x)   (x) = w (x0 )

(x0 xi )w (xi )
x xi
i=1
i=1 0

y2
y3
y1
+
+
= w (2)
(2 3)w (3) (2 5)w (5) (2 6)w (6)

1
1
1
+
+
.
+y0
23 25 26

415

## w (2) = 12, w (3) = 6, w (5) = 6, w (6) = 12.

Thus

34
136
229
1 1

+
+
+ 13 1
= 14.
f (2)  12
(1) 6 (3) (6) (4) 12
3 4
Also
f  (2.5)
 w (2.5)

3

i=0


yi
yi

w(2.5)

(2.5 xi )w (xi )
(2.5 xi )2 w (xi )
3

i=0

y1
y2
y3
y0

+
+
+
= w (2.5)
(2.5 2)w (2) (2.5 3)w (3) (2.5 5)w (5) (2.5 6)w (6)

y0
y1
y2
y3
w(2.5)
+
+
+
(2.52)2 w (2) (2.53)2 w (3) (2.55)2 w (5) (2.56)2 w (6)
Now, w (2.5) = 1.5, w(2.5) = 2.1875.
Therefore,

34
136
229
13

+
+
+
f (2.5)  1.5
0.5 (12) (0.5) 6 (2.5) (6) (3.5) 12

34
136
229
13
+
+
+
+ 2.1875
(0.5)2(12) (0.5)26 (2.5)2(6) (3.5)212
= 20.75.

Algorithm 7.1 (Derivative). This algorithm determines the rst order derivative
of a function given in tabular form (xi , yi ), i = 0, 1, 2, . . . , n, at a given point xg, xg
may or may not be equal to the given nodes xi , based on Lagranges interpolation.
Algorithm Derivative Lagrange
Read xi , yi , i = 0, 1, 2, . . . , n.
Read xg; //the point at which the derivative is to be evaluated.//
Compute w (xj ), j = 0, 1, 2, . . . , n, using the function wd(j).
Set sum1 = sum2 = 0;
Check xg is equal to given nodes xi .
If xg is not equal to any node then
for i = 0 to n do
Compute t = yi /((xg xi ) wd(i));
Compute sum1 = sum1 + t;
Compute sum2 = sum2 + t/(xg xi );

## 416 Numerical Analysis

endfor;
//compute w (xg)//
set t = 0;
for j = 0 to n do
set prod = 1;
for i = 0 to n do
if (i = j) then prod = prod (xg xi );
endfor;
Compute t = t + prod;
endfor;
//compute w(xg) //
set t1 = 1;
for i = 0 to n do
Compute t1 = t1 (xg xi );
endfor;
Compute result = t sum1 t1 sum2;
else //xg is equal to xj //
for i = 0 to n do
if i = j then
Compute sum1 = sum1 + yi /((xj xi ) wd(i));
Compute sum2 = sum2 + 1/(xj xi );
endif;
endfor;
Compute result = wd(j) sum1 + yj sum2;
endif;
Print The value of the derivative, result;
function wd(j)
//This function determines w (xj ).//
Set prod = 1;
for i = 0 to n do
if (i = j) prod = prod xi ;
endfor;
return prod;
end wd
end Derivative Lagrange
Program 7.1
.
/* Program Derivative
Program to find the first order derivative of a function
y=f(x) given as (xi,yi), i=0, 1, 2, ..., n, using formula
based on Lagranges interpolation. */
#include<stdio.h>

## int n; float x,xg;

float wd(int);
void main()
{
int i,j,flag=-1;
float y,sum1=0,sum2=0,prod,t,t1=1,result;
printf("Enter the value of n and the data in the form
(x[i],y[i]) \n");
scanf("%d",&n);
for(i=0;i<=n;i++) scanf("%f %f",&x[i],&y[i]);
printf("Enter the value of x at which derivative
is required \n");
scanf("%f",&xg);
for(i=0;i<=n;i++) if(x[i]==xg) flag=i;
if(flag==-1) /* xg is not equal to xi, i=0, 1, ..., n */
{
for(i=0;i<=n;i++)
{
t=y[i]/((xg-x[i])*wd(i));
sum1+=t;
sum2+=t/(xg-x[i]);
}
/* Computation of w(xg) */
t=0;
for(j=0;j<=n;j++)
{
prod=1;
for(i=0;i<=n;i++) if(i!=j) prod*=(xg-x[i]);
t+=prod;
}
/* computation of w(xg) */
for(i=0;i<=n;i++) t1*=(xg-x[i]);
result=t*sum1-t1*sum2;
} /* end of if part */
else
{
j=flag;
for(i=0;i<=n;i++)
if(i!=j)

417

## 418 Numerical Analysis

{
sum1+=y[i]/((x[j]-x[i])*wd(i));
sum2+=1/(x[j]-x[i]);
}
result=wd(j)*sum1+y[j]*sum2;
} /* end of else part */
printf("The value of derivative at x= %6.4f is
%8.5f",xg,result);
}
/* this function determines w(xj) */
float wd(int j)
{
int i;float prod=1;
for(i=0;i<=n;i++) if(i!=j) prod*=(x[j]-x[i]);
return prod;
}
A sample of input/output:
Enter the value of n and the data in the form (x[i],y[i])
3
1 0.54030
2 -0.41615
3 -0.98999
4 -0.65364
Enter the value of x at which derivative is required
1.2
The value of derivative at x= 1.2000 is -0.99034
Table of derivatives
The summary of the formulae of derivatives based on nite dierences.


1
1
1
1
1
1
2 + 3 4 + 5 6 + y0
h
2
3
4
5
6


1
7 7 363 8

2
3 11 4 5 5 137 6
+
+ y0
f (x0 ) 2 + +
h
12
6
180
10
560


1
1
1
1
1
1
+ 2 + 3 + 4 + 5 + 6 + yn
f  (xn ) 
h
2
3
4
5
6


1

2
3 11 4 5 5 137 6 7
7 363 8
+ +
+ yn
f (xn )  2 + + + +
h
12
6
180
10
560
f  (x0 ) 

(7.32)
(7.33)
(7.34)
(7.35)

## Differentiation and Integration



1 y1+y0 3 y2+3 y1 5 y3+5 y2

+
+ y0
f (x0 ) 
h
2
12
60


1
1 4
1 6

2
f (x0 )  2 y1 y2 + y3 y0
h
12
90


7.6

419

(7.36)
(7.37)

## Two-point and Three-point Formulae

Only the rst term of (7.32), gives a simple formula for the rst order derivative
f  (xi ) 

yi+1 yi
y(xi + h) y(xi )
yi
=
=
.
h
h
h

(7.38)

## Similarly, the equation (7.34), gives

f  (xi ) 

yi yi1
y(xi ) y(xi h)
yi
=
=
.
h
h
h

(7.39)

Adding equations (7.38) and (7.39), we obtain the central dierence formula for rst
order derivative, as
f  (xi ) 

y(xi + h) y(xi h)
.
2h

(7.40)

## Equations (7.38)-(7.40) give two-point formulae to nd rst order derivative at x = xi .

Similarly, from equation (7.33)
f  (xi ) 

2 yi
yi+2 2yi+1 + yi
y(xi +2h) 2y(xi + h)+y(xi )
=
=
.
2
2
h
h
h2

(7.41)

## From equation (7.35)

f  (xi ) 

2 yi
yi 2yi1 +yi2
y(xi )2y(xi h)+y(xi 2h)
=
=
.
2
2
h
h
h2

(7.42)

## Equation (7.37) gives

f  (x0 ) 

2 y1
y1 2y0 + y1
y(x0 + h) 2y(x0 ) + y(x0 h)
=
=
.
2
2
h
h
h2

In general,
f  (xi ) 

## y(xi + h) 2y(xi ) + y(xi h)

.
h2

(7.43)

Equations (7.41)-(7.43) give the three-point formulae for second order derivative.

7.6.1

## Error analysis and optimum step size

The truncation error of the two-point formula (7.40) is O(h2 ). Assume that f
C 3 [a, b] (i.e., f is continuously dierentiable up to third order within [a, b]) and x
h, x, x + h [a, b]. Then by Taylors series
h2 
f (xi ) +
2!
h2
and f (xi h) = f (xi ) hf  (xi ) + f  (xi )
2!
f (xi + h) = f (xi ) + hf  (xi ) +

h3 
f (1 )
3!
h3 
f (2 )
3!

By subtraction
f (xi + h) f (xi h) = 2hf  (xi ) +

f  (1 ) + f  (2 ) 3
h .
3!

(7.44)

Since f  is continuous, by the intermediate value theorem there exist a number so
that
f  (1 ) + f  (2 )
= f  ().
2
Thus, after rearrangement the equation (7.44) becomes
f  (xi ) =

## f (xi + h) f (xi h) f  ()h2

.
2h
3!

(7.45)

It may be noted that the rst term of right hand side is two-point formula while
second term is the truncation error and it is of O(h2 ).
To nd the computers round-o error, it is assumed that f (x0 h) = y(x0 h)+1
and f (x0 + h) = y(x0 + h) + 1 where y(x0 h) and y(x0 + h) are the approximate
values of the original function f at the points (x0 h) and (x0 + h) respectively and
1 and 1 are the round-o errors.
Thus
f  (xi ) =

y(xi + h) y(xi h)
+ Etrunc
2h

and
y(xi + h) y(xi h)
+ Etrunc + Eround
2h
y(xi + h) y(xi h)
+E
=
2h

f  (xi ) =

where
E = Eround + Etrunc =

1 1 h2 f  ()

2h
6

(7.46)

## Differentiation and Integration

421

is the total error accumulating the round-o error (Eround ) and the truncation error
(Etrunc ).
Let |1 | , |1 | and M3 = max |f  (x)|.
axb

Then from (7.46), the upper bound of the total error is given by
|E|

M3 h2
|1 | + |1 | h2 
+ |f ()| +
.
2h
6
h
6

## Now, |E| will be minimum for a given h if

optimum value of h to minimize |E| is

h=

3
M3

(7.47)

d|E|
hM3
= 0 i.e., 2 +
= 0. Thus the
dh
h
3
1/3
(7.48)



3
|E| =
M3

1/3

M3
+
6

3
M3

2/3
.

## Example 7.6.1 If f C 5 [a, b] (i.e., the function f is continuously dierentiable up

to fth order on [a, b]) and x 2h, x h, x + h, x + 2h [a, b], then show that
f  (x) 

f (x + 2h) + 8f (x + h) 8f (x h) + f (x 2h) h4 f v ()
+
,
12h
30

(7.49)

where lies between x 2h and x + 2h. Determine the optimal value of h when
(i) |Eround | = |Etrunc |,
(ii) total error |Eround | + |Etrunc | is minimum.
Solution. By Taylors series expansion for step length h and h,
f (x + h) = f (x) + hf  (x) +

h2 
h3
h4
h5
f (x) + f  (x) + f iv (x) + f v (1 )
2!
3!
4!
5!

f (x h) = f (x) hf  (x) +

h2 
h3
h4
h5
f (x) f  (x) + f iv (x) f v (2 ).
2!
3!
4!
5!

and

Then by subtraction
f (x + h) f (x h) = 2hf  (x) +

+
.
3!
5!

## 422 Numerical Analysis

Similarly, when step size is 2h then
f (x + 2h) f (x 2h) = 4hf  (x) +

+
.
3!
5!

## All s lie between x 2h and x + 2h.

Now,
f (x 2h) f (x + 2h) + 8f (x + h) 8f (x h)
16f v (3 ) 64f v (4 ) 5
h .
= 12hf  (x) +
120
Since f v (x) is continuous, f v (3 )  f v (4 ) = f v () (say).
Then 16f v (3 ) 64f v (4 ) = 48f v ().
Using this result the above equation becomes
2
f (x 2h) f (x + 2h) + 8f (x + h) 8f (x h) = 12hf  (x) h5 f v ().
5
Hence, the value of f  (x) is given by
f  (x) 

## f (x + 2h) + 8f (x + h)8f (x h)+f (x 2h) f v ()h4

+
.
12h
30

(7.50)

The rst term on the right hand side is a four-point formula to nd f  (x) and second
term is the corresponding truncation error.
Let f (x+2h) = y2 +2 , f (x+h) = y1 +1 , f (xh) = y1 +1 and f (x2h) = y2 +
2 , where yi and i are the approximate values of f (x + ih) and the corresponding
round-o errors respectively. Also let
max
|f v ()|.
M5 =
x2hx+2h

## Then (7.50) becomes

y2 + 8y1 8y1 + y2 2 + 81 81 + 2 f v ()h4
+
+
12h
12h
30
y2 + 8y1 8y1 + y2
+ Eround + Etrunc .
=
12h

f  (x) =

## Let = max{|2 |, |1 |, |1 |, |2 |}. Then

|Eround |
and |Etrunc |

M5 h4
.
30

18
3
|2 | + 8|1 | + 8|1 | + |2 |

=
12h
12h
2h

## Differentiation and Integration

423

3
45
M5 h4
.
=
or, h4 =
2h
30
M5
1/4

45
and
Thus the optimum value of h is
M5


 9 1/4
3 M5 1/4
= (M5 3 )1/4
.
|Eround | = |Etrunc | =
2 45
80
(i) If |Eround | = |Etrunc | then

or,

3
4M5 h3
45
= 0, i.e., h5 =
+
.
2
2h
30
4M5

45
4M5

d|E|
=0
dh

1/5
.

f  (x0 ) =

+
,
12h
30

(7.51)

## where x2 < < x2 .

Example 7.6.2 The value of x and f (x) = x cos x are tabulated as follows:
x :
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
f (x) : 0.19601 0.28660 0.36842 0.43879 0.49520 0.53539 0.55737 0.55945 0.54030

Find the value of f  (0.6) using the two- and four-point formulae
f (x0 + h) f (x0 h)
2h
f (x0 + 2h) + 8f (x0 + h) 8f (x0 h) + f (x0 2h)
and
f  (x0 ) =
12h
with step size h = 0.1.
f  (x0 ) =

## Solution. By the two-point formula,

0.53539 0.43879
f (0.7) f (0.5)
=
= 0.48300
f  (0.6) 
0.2
0.2
and by the four-point formula,
f (0.8) + 8f (0.7) 8f (0.5) + f (0.4)
12 0.1
0.55737 + 8 0.53539 8 0.43879 + 0.36842
=
= 0.48654.
1.2

f  (0.6) 

## 424 Numerical Analysis

The exact value is f  (0.6) = cos 0.6 0.6 sin 0.6 = 0.48655.
Therefore, error in two-point formula is 0.00355 and that in four-point formula is
0.00001. Clearly, four-point formula gives better result than two-point formula.

7.7

## Richardsons Extrapolation Method

The improvement of derivative of a function can be done using this method. This
method reduces the number of function evaluation to achieve the higher order accuracy.
The formula to nd the rst order derivative using two points is
f  (x) =

f (x + h) f (x h)
+ Etrunc = g(h) + Etrunc
2h

where Etrunc is the truncation error and g(h) is the approximate rst order derivative
of f (x).
Using Taylors series expansion, it can be shown that, Etrunc is of the following form.
Etrunc = c1 h2 + c2 h4 + c3 h6 + .
The Richardsons extrapolation method combines two values of f  (x) obtained
by a certain method with two dierent step sizes, say, h1 and h2 . Generally, h1 and h2
are taken as h and h/2. Thus
f  (x) = g(h) + Etrunc = g(h) + c1 h2 + c2 h4 + c3 h6 +
h2
h4
h6
and f  (x) = g(h/2) + c1 + c2 + c3 + .
4
16
32

(7.52)
(7.53)

## The constants c1 , c2 , . . . are independent of h. Eliminating c1 between (7.52) and

(7.53), and we get
4g(h/2) g(h) 3
7
c2 h4 c3 h6 +
3
4
8
4g(h/2) g(h)
+ d1 h4 + d2 h6 +
=
3

f  (x) =

(7.54)

Denoting
4g(h/2) g(h)
3

by

g1 (h/2),

(7.55)

## equation (7.54) becomes

f  (x) = g1 (h/2) + d1 h4 + O(h6 ).

(7.56)

## Differentiation and Integration

425

This equation shows that g1 (h/2) is an approximate value of f  (x) with fourth-order
accuracy. Thus a result accurate up to fourth order is obtained by combining two results
accurate up to second order.
Now, by repeating the above result one can obtain
f  (x) = g1 (h/2) + d1 h4 + O(h6 )
h4
f  (x) = g1 (h/22 ) + d1 + O(h6 )
16

(7.57)

## Eliminating d1 from (7.57) to nd O(h6 ) order formula, as

f  (x) = g2 (h/22 ) + O(h6 ),

(7.58)

where
g2 (h/22 ) =

42 g1 (h/22 ) g1 (h/2)
.
42 1

(7.59)

## Thus g2 (h/22 ) is the sixth-order accurate result of f  (x).

Hence the successive higher order results can be obtained from the following formula
 


h
h
  4k gk1 m gk1 m1
h
2
2
,
(7.60)
gk m =
k
2
4 1
k = 1, 2, 3, . . . ; m = k, k + 1, . . .
where g0 (h) = g(h).
This process is called repeated extrapolation to the limit. The values of gk (h/2m ) for
dierent values of k and m are tabulated as shown in Table 7.1.
Table 7.1: Richardsons extrapolation table
h
h

second
order
g(h)

fourth
order

sixth
order

eight
order

g1 (h/2)
g2 (h/22 )

h/2 g(h/2)
g1
h/22

(h/22 )

g(h/22 )

g2
g1

h/23 g(h/23 )

g3 (h/23 )

(h/23 )

(h/23 )

## 426 Numerical Analysis

It may be noted that the successive values in a particular column give better approximations of the derivative than the preceding columns. This process will terminate
when
|gm (h/2) gm1 (h)|
for a given error tolerance .
We have seen that one approximate value of f  (x) is g1 (h/2) where
g1 (h/2) =

g(h/2) g(h)
4g(h/2) g(h)
= g(h/2) +
.
3
3

Here g(h/2) is more accurate than g(h) and then g1 (h/2) gives an improved approximation over g(h/2). If g(h) < g(h/2), g1 (h/2) > g(h/2) and if g(h/2) < g(h),
g1 (h/2) < g(h/2). Thus the value of g1 (h/2) lies outside the interval [g(h), g(h/2)] or
[g(h/2), g(h)] as the case may be. Thus g1 (h/2) is obtained from g(h) and g(h/2) by
means of an extrapolation operation. So, this process is called (Richardson) extrapolation.
Example 7.7.1 Use Richardsons extrapolation method to nd f  (0.5) where
f (x) = 1/x starting with h = 0.2.
Solution. Here h = 0.2 and x = 0.5.
1
1
Then

f (x + h) f (x h)
= 0.5 + 0.2 0.5 0.2 = 4.76190,
g(h) =
2h
2 0.2
1
1

f (x + h/2) f (x h/2)
g(h/2) =
= 0.5 + 0.1 0.5 0.1 = 4.16667.
2(h/2)
0.2
4g(h/2) g(h)
4 (4.16667)(4.76190)
=
= 3.96826.
41
3
Halving the step size further, we compute

Then g1 (h/2) =

1
1

## g(h/22 ) = 0.5 + 0.05 0.5 0.05 = 4.04040,

2 0.05
2 ) g(h/2)
4g(h/2
4 (4.04040) (4.16667)
g1 (h/22 ) =
=
= 3.99831.
41
3
Now,

16 (3.99831) (3.96826)
42 g1 (h/22 ) g1 (h/2)
=
2
4 1
15
= 4.00031.

g2 (h/22 ) =

427

## The above calculations are tabulated as follows:

h
g
0.2 4.76190

g1

g2

3.96826
0.1 4.16667

4.00031
3.99831

0.05 4.04040
Thus, after two steps we found that f  (0.5) = 4.00031 while the exact value is

1
= 4.0.
f  (0.5) = 2
x x=0.5
Example 7.7.2 For the following table
x : 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
f (x) : 1 2/3 1/2 2/5 1/3 2/7 1/4 2/9 1/5 2/11 1/6
nd the value of f  (3) using Richardsons extrapolation.
Solution. Let x = 3 and h = 2.
1 1

f (5) f (1)
f (x + h) f (x h)
=
= 6 2 = 0.083333.
g(h) =
2h
22
4
1 1

f (x + h/2) f (x h/2)
f (4) f (2)
5
3 = 0.066666.
g(h/2) =
=
=
2(h/2)
2
2
4 (0.06666) (0.083333)
4g(h/2) g(h)
=
= 0.061110.
g1 (h/2) =
41
3
2 2

f (3.5) f (2.5)
2
9
7 = 0.063492.
g(h/2 ) = g(0.5) =
=
2 0.5
1
4 (0.063492) (0.066666)
4g(h/22 ) g(h/2)
=
= 0.062434.
g1 (h/22 ) =
41
3
Thus
42 g1 (h/22 ) g1 (h/2)
42 1
16 (0.062434) (0.061110)
= 0.062522.
=
15

g2 (h/22 ) =

## Algorithm 7.2 (Richardsons extrapolation). This algorithm is used to nd

the derivative using Richardsons extrapolation formula.
The formula (7.60) can be written as
gk (h) =

4k 1

## at each iteration h becomes half of its previous value.

We denote gk1 (2h) (calculated as previous iteration) by g0 (k 1) (old value) and
gk1 (h) by gn (k 1) (new value) to remove h from the formula.
Then the above formula reduces to
4k gn (k 1) go (k 1)
, k = 1, 2, . . .
4k 1
f (x + h) f (x h)
= g0 (0).
where gn (0) =
2h
gn (k) =

## Algorithm Richardson extrapolation

Input function f (x);
Read x, h, ; //error tolerance//
Compute g0 (0) =
Set j = 1;
10: Set h = h/2;
Compute gn (0) =

f (x + h) f (x h)
;
2h
f (x + h) f (x h)
;
2h

for k = 1 to j do
4k gn (k 1) go (k 1)
;
4k 1
if |gn (j 1) gn (j)| < then
Print gn (j) as the value of derivative;
Stop;
else
for k = 0 to j do
g0 (k) = gn (k); //set new values as old values//
j = j + 1;
goto 10;
endif;
end Richardson extrapolation
gn (k) =

## Differentiation and Integration

Program 7.2
.
/* Program Richardson Extrapolation
This program finds the first order derivative of a function
f(x) at a given value of x by Richardson Extrapolation.
Here we assume that f(x)=1/(x*x).
*/
#include<stdio.h>
#include<math.h>
#include<stdlib.h>
void main()
{
int j=1,k;
float x,h,eps=1e-5,g0,gn;
float f(float x);
printf("Enter the value of x ");
scanf("%f",&x);
printf("Enter the value of h ");
scanf("%f",&h);
g0=(f(x+h)-f(x-h))/(h+h);
start: h=h/2;
gn=(f(x+h)-f(x-h))/(h+h);
for(k=1;k<=j;k++)
gn[k]=(pow(4,k)*gn[k-1]-g0[k-1])/(pow(4,k)-1);
if(fabs(gn[j-1]-gn[j])<eps)
{
printf("The derivative is %8.5f at %8.5f ",gn[j],x);
exit(0);
}
else
{
for(k=0;k<=j;k++) g0[k]=gn[k];
j++;
goto start;
}
} /* main */
/* definition of f(x) */
float f(float x)
{
return(1/(x*x));
}

429

## 430 Numerical Analysis

A sample of input/output:
Enter the value of x 1.5
Enter the value of h 0.5
The derivative is -0.59259 at

7.8

1.50000

## Cubic Spline Method

The cubic spline may be used to determine the rst and second derivatives of a function.
This method works into two stages. In rst stage the cubic splines will be constructed
with suitable intervals and in second stage the rst and second derivatives are to be
determined from the appropriate cubic spline. This method is labourious than the
other methods, but, once a cubic spline is constructed then the method becomes very
ecient. The process of nding derivative is illustrated by as example in the following.
Example 7.8.1 Let y = f (x) = cos x, 0 x /2 be the function. Find the natural cubic spline in the intervals 0 x /4 and /4 x /2 and hence determine
the approximate values of f  (/8) and f  (/8). Also, use two-point formula to nd
the value of f  (/8). Find the error in each case.

## Solution. Here n = 2. Therefore, h = /4, y0 = cos 0 = 1, y1 = cos /4 = 1/ 2 and

y2 = cos /2 = 0.
Also, M0 = M2 = 0.
Then by the formula (3.99)
M0 + 4M1 + M2 =

6
[y0 2y1 + y2 ]
h2

96
24
That is, 4M1 = 2 (1 2) or, M1 = 2 (1 2) = 1.007246602.

s1 (x), 0 x /4
S(x) =
s2 (x), /4 x /2,
where
s1 (x) =
and



4 x3
1

2
M1 1 + M1 x +
,
6
4
2 96


 1
2
4 (/2 x)3
M1 M1 (/2 x) .
s2 (x) =

6
2 96

## Differentiation and Integration

431

Two-point formula.
Let h = /30.
f (/8 + /30) f (/8 /30)
= 0.381984382.
Then f  (/8) 
2./30
The actual value of f  (/8) is sin /8 = 0.382683432.
Therefore, error in cubic spline method is 0.042722272 while in two-point formula
that is 0.000699050365.

7.9

## Determination of Extremum of a Tabulated Function

It is known that, if a function is dierentiable then the maximum and minimum value
of that function can be determined by equating the rst derivative to zero and solving
for the variable. The same method is applicable for the tabulated function.
Now, consider the Newtons forward dierence interpolation formula.
y = f (x) = y0 + uy0 +
where u =
Then

u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0 + ,
2!
3!

(7.61)

x x0
.
h

2u 1 2
3u2 6u + 2 3
dy
= y0 +
y0 +
y0 + .
dx
2
6
dy
= 0. Then
For maxima and minima
dx
y0 +

2u 1 2
3u2 6u + 2 3
y0 +
y0 + = 0
2
6

(7.62)

For simplicity, the third and higher dierences are neglected and obtain the quadratic
equation for u as
au2 + bu + c = 0,

(7.63)

1
1
1
where a = 3 y0 , b = 2 y0 3 y0 , c = y0 2 y0 + 3 y0 .
2
2
3
The values of u can be determined by solving this equation. Then the values of x
are determined from the relation x = x0 + uh. Finally, the maximum value of y can be
obtained from the equation (7.61).
Example 7.9.1 Find x for which y is maximum and also nd the corresponding
value of y, from the following table.
x :
1.5
2.0
2.5
3.0
3.5
y : 0.40547 0.69315 0.91629 1.09861 1.25276

## 432 Numerical Analysis

Solution. The forward dierence table is
x
y
1.5 0.40547

2 y

0.28768
2.0 0.69315

0.06454
0.22314

2.5 0.91629

0.04082
0.18232

3.0 1.09861

0.02817
0.15415

3.5 1.25276
Let x0 = 1.5. Using formula (7.62) we have
2u 1 2
2u 1
y0 = 0 or, 0.28768 +
(0.06454) = 0 or, u = 4.95739.
y0 +
2
2
Therefore, x = x0 + uh = 1.5 + 0.5 4.95739 = 3.97870.
For this x, the value of y is obtained by Newtons backward formula as
y(3.97870) = 1.25276 + 0.47870 (0.15415)
0.47870(0.47870 + 1)
+
(0.02817) = 1.31658.
2
This is the approximate maximum value of y when x = 3.97870.

7.10

Integration

It is well known that, if a function f (x) is known completely, even then it is not always
possible to evaluate the denite integral of it using analytic method. Again, in many
real life problems, we are required to integrate a function between two given limits,
but the function is not known explicitly, but, it is known in a tabular form (equally or
unequally spaced). Then a method, known as numerical integration or quadrature
can be used to solve all such problems.
The problem of numerical integration is stated below:
Given a set of data points (x0 , y0 ), (x1 , y1 ), . . . , (xn , yn ) of a function y = f (x), it is
8b
required to nd the value of the denite integral a f (x) dx. The function f (x) is replaced
by a suitable interpolating polynomial (x).
Then the approximate value of the denite integral is calculated using the following
formula
% b
% b
f (x) dx 
(x) dx.
(7.64)
a

## Differentiation and Integration

433

Thus, dierent integration formulae can be derived depending on the type of the
interpolation formulae used.
A numerical integration formula is said to be of closed type, if the limits of integration a and b are taken as interpolating points. If a and b are not taken as interpolating
points then the formula is known as open type formula.

7.11

Interpolation

## The Newtons forward interpolation formula for the equispaced points xi , i = 0, 1, . . . , n,

xi = x0 + ih is
u(u 1) 2
u(u 1)(u 2) 3
y0 +
y0 + ,
2!
3!

(x) = y0 + uy0 +

(7.65)

x x0
, h is the spacing.
h
Let the interval [a, b] be divided into n equal subintervals such that a = x0 < x1 <
x2 < < xn = b. Then
% xn
% b
f (x) dx 
(x) dx
I=
where u =

x0

a
xn

=
x0



u2 u 2
u3 3u2 + 2u 3
y0 +
y0 + dx.
y0 + uy0 +
2!
3!

## Since x = x0 + uh, dx = h du, when x = x0 then u = 0 and when x = xn then u = n.

Thus,
% n

u2 u 2
u3 3u2 + 2u 3
I=
y0 + uy0 +
y0 +
y0 + hdu
2!
3!
0

 u2 n 2 y  u3 u2 n 3 y  u4
n
0
0
n
3
2

u +u
= h y0 [u]0 + y0
+
+
+
2 0
2!
3
2 0
3!
4
0


n
2n2 3n 2
n3 4n2 + 4n 3
y0 +
y0 + .
= nh y0 + y0 +
(7.66)
2
12
24
From this formula, one can generate dierent integration formulae by substituting
n = 1, 2, 3, . . . .
7.11.1

Trapezoidal Rule

Substituting n = 1 in the equation (7.66). In this case all dierences higher than the
rst dierence become zero. Then
% xn



 h
1
1
f (x) dx = h y0 + y0 = h y0 + (y1 y0 ) = (y0 + y1 ).
(7.67)
2
2
2
x0

## 434 Numerical Analysis

The formula (7.67) is known as the trapezoidal rule.
In this formula, the interval [a, b] is considered as a single interval, and it gives a
very rough answer. But, if the interval [a, b] is divided into several subintervals and this
formula is applied to each of these subintervals then a better approximate result may
be obtained. This formula is known as composite formula, deduced below.
Composite trapezoidal rule
Let the interval [a, b] be divided into n equal subintervals by the points a = x0 , x1 , x2 ,
. . ., xn = b, where xi = x0 + ih, i = 1, 2, . . . , n.
Applying the trapezoidal rule to each of the subintervals, one can nd the composite
formula as
% x1
% x2
% xn
% b
f (x) dx =
f (x) dx +
f (x) dx + +
f (x) dx
x0

x1

xn1

h
h
h
h
 [y0 + y1 ] + [y1 + y2 ] + [y2 + y3 ] + + [yn1 + yn ]
2
2
2
2
h
= [y0 + 2(y1 + y2 + + yn1 ) + yn ].
(7.68)
2
Error in trapezoidal rule
The error of trapezoidal rule is
%
E=
a

f (x) dx

h
(y0 + y1 ).
2

(7.69)

Let y = f (x) be continuous and possesses continuous derivatives of all orders. Also,
it is assumed that there exists a function F (x) such that F  (x) = f (x) in [x0 , x1 ].
Then
% x1
% b
f (x) dx =
F  (x) dx = F (x1 ) F (x0 )
a

x0

## = F (x0 + h) F (x0 ) = F (x0 ) + hF  (x0 ) +

h3 
F (x0 ) + F (x0 )
3!
h2
h3
= hf (x0 ) + f  (x0 ) + f  (x0 ) +
2!
3!
h2 
h3 
= hy0 + y0 + y0 +
2
6

h2 
F (x0 )
2!

(7.70)

## Differentiation and Integration

435

Again,
h
h
(y0 + y1 ) = [y0 + y(x0 + h)]
2
2
h
h2
= [y0 + y(x0 ) + hy  (x0 ) + y  (x0 ) + ]
2
2!
2
h
h
= [y0 + y0 + hy0 + y0 + ].
2
2!

(7.71)

## Using (7.70) and (7.71), equation (7.69) becomes


 h

h
h2
h2
E = h y0 + y0 + y0 + 2y0 + hy0 + y0 +
2
6
2
2!
h3 
= y0 +
12
h3 
h3
= f (x0 ) +  f  (),
12
12

(7.72)

## where a = x0 < < x1 = b.

Equation (7.72) gives the error in the interval [x0 , x1 ].
The total error in the composite rule is
E=

h3 

(y + y1 + + yn1
).
12 0


then
If y  () is the largest among the n quantities y0 , y1 , . . . , yn1

(b a) 2 
1 3 
h ny () =
h y (), as nh = b a.
12
12

Note 7.11.1 The error term shows that if the second and higher order derivatives of
f (x) vanish then the trapezoidal rule gives exact result of the integral. This means, the
method gives exact result when f (x) is linear.
Geometrical interpretation of trapezoidal rule
In this rule, the curve y = f (x) is replaced by the line joining the points A(x0 , y0 ) and
B(x1 , y1 ) (Figure 7.1). Thus the area bounded by the curve y = f (x), the ordinates
x = x0 , x = x1 and the x-axis is then approximately equivalent to the area of the
trapezium (ABCD) bounded by the line AB, x = x0 , x = x1 and x-axis.
The geometrical signicance of composite trapezoidal rule is that the curve y = f (x)
is replaced by n straight lines joining the points (x0 , y0 ) and (x1 , y1 ); (x1 , y1 ) and (x2 , y2 );
. . ., (xn1 , yn1 ) and (xn , yn ). Then the area bounded by the curve y = f (x), the lines
x = x0 , x = xn and the x-axis is then approximately equivalent to the sum of the area
of n trapeziums (Figures 7.2).

y
6

y = f (x)
B

A
y1

y0
O D x0

C x1

- x

y
6

x0

x1

x2

xn

- x

## Alternative deduction of trapezoidal rule

Let f C 2 [a, b], where [a, b] is a nite interval. Now, transfer the interval [a, b] to [1, 1]
a+b ba
+
t = p + qt (say).
using the relation x =
2
2
Let f (x) = f (p + qt) = g(t). When x = a, b then t = 1, 1, i.e., g(1) = f (b), g(1) =
f (a).
Thus
%

I=

[g(t) + g(t)]dt.

=q
0

g(t) q dt = q
1

%

f (x)dx =

g(t)dt +

g(t)dt
0

437

## Now, applying integration by parts.



1

q
t[g  (t) g  (t)]dt
0
0
% 1
t.2tg  (c)dt, where 0 < c < 1
= q[g(1) + g(1)] q

I = q {g(t) + g(t)}t



## [by MVT of integral calculus]

2
= q[f (a) + f (b)] qg  (d)
3
2 3 
= q[f (a) + f (b)] q f (p + qd)
3
2 3 
= q[f (a) + f (b)] q f (), where a < < b
3
2  b a 3 
ba
[f (a) + f (b)]
f ()
=
2
3
2
1
h
= [f (a) + f (b)] h3 f  (), as h = b a.
2
12
In this expression, the rst term is the approximate integration obtained by trapezoidal rule and the second term represents the error.
% b
Algorithm 7.3 (Trapezoidal). This algorithm nds the value of
f (x)dx based
a

## on the tabulated values (xi , yi ), yi = f (xi ), i = 0, 1, 2, . . . , n, using trapezoidal rule.

Algorithm Trapezoidal
Input function f (x);
Read a, b, n; //the lower and upper limits and number of subintervals.//
Compute h = (b a)/n;
1
Set sum = [f (a) + f (a + nh)];
2
for i = 1 to n 1 do
Compute sum = sum + f (a + ih);
endfor;
Compute result = sum h;
Print result;
end Trapezoidal

## 438 Numerical Analysis

Program 7.3
.
/* Program Trapezoidal
This program finds the value of integration of a function
by trapezoidal rule.
Here we assume that f(x)=x^3. */
#include<stdio.h>
void main()
{
float a,b,h,sum; int n,i;
float f(float);
printf("Enter the values of a, b ");
scanf("%f %f",&a,&b);
printf("Enter the value of n ");
scanf("%d",&n);
h=(b-a)/n;
sum=(f(a)+f(a+n*h))/2.;
for(i=1;i<=n-1;i++) sum+=f(a+i*h);
sum=sum*h;
printf("The value of the integration is %8.5f ",sum);
}
/* definition of the function f(x) */
float f(float x)
{
return(x*x*x);
}
A sample of input/output:
Enter the values of a, b 0 1
Enter the value of n 100
The value of the integration is
7.11.2

0.25002

## Simpsons 1/3 rule

In this formula the interval [a, b] is divided into two equal subintervals by the points
x0 , x1 , x2 , where h = (b a)/2, x1 = x0 + h and x2 = x1 + h.
This rule is obtained by putting n = 2 in (7.66). In this case, the third and higher
order dierences do not exist.

439

## The equation (7.66) is simplied as

% xn
1 2
1
f (x) dx  2h y0 + y0 + y0 = 2h[y0 + (y1 y0 ) + (y2 2y1 + y0 )]
6
6
x0
h
= [y0 + 4y1 + y2 ].
(7.73)
3
The above rule is known as Simpsons 1/3 rule or simply Simpsons rule.
Composite Simpsons 1/3 rule
Let the interval [a, b] be divided into n (an even number) equal subintervals by the
points x0 , x1 , x2 , . . . , xn , where xi = x0 + ih, i = 1, 2, . . . , n. Then
% b
% x2
% x4
% xn
f (x) dx =
f (x) dx +
f (x) dx + +
f (x) dx
x0

x2

xn2

h
h
h
= [y0 + 4y1 + y2 ] + [y2 + 4y3 + y4 ] + + [yn2 + 4yn1 + yn ]
3
3
3
h
= [y0 + 4(y1 + y3 + + yn1 ) + 2(y2 + y4 + + yn2 ) + yn ].
3
(7.74)
This formula is known as Simpsons 1/3 composite rule for numerical integration.
Error in Simpsons 1/3 rule
The error in this formula is

xn

E=
x0

f (x) dx

h
[y0 + 4y1 + y2 ].
3

(7.75)

Let the function f (x) be continuous in [x0 , x2 ] and possesses continuous derivatives
of all order. Also, let there exists a function F (x) in [x0 , x2 ], such that F  (x) = f (x).
Then
% x2
% x2
f (x) dx =
F  (x) dx = F (x2 ) F (x0 )
x0

x0

## = F (x0 + 2h) F (x0 ) = F (x0 ) + 2hF  (x0 ) +

(2h)2 
F (x0 )
2!

(2h)3 
(2h)4 iv
(2h)5 v
F (x0 ) +
F (x0 ) +
F (x0 ) + F (x0 )
3!
4!
5!
4
2
= 2hf (x0 ) + 2h2 f  (x0 ) + h3 f  (x0 ) + h4 f  (x0 )
3
3
4 5 iv
+ h f (x0 ) + .
(7.76)
15
+

## 440 Numerical Analysis

Again,
h
h
[y0 + 4y1 + y2 ] = [f (x0 ) + 4f (x1 ) + f (x2 )]
3
3
h
= [f (x0 ) + 4f (x0 + h) + f (x0 + 2h)]
3

h2
h3
h
f (x0 ) + 4 f (x0 ) + hf  (x0 ) + f  (x0 ) + f  (x0 )
=
3
2!
3!


4
h
(2h)2 
+ f iv (x0 ) + + f (x0 ) + 2hf  (x0 ) +
f (x0 )
4!
2!


(2h)3 
(2h)4 iv
+
f (x0 ) +
f (x0 ) +
3!
4!
4
2
= 2hf (x0 ) + 2h2 f  (x0 ) + h3 f  (x0 ) + h4 f  (x0 )
3
3
5
(7.77)
+ h5 f iv (x0 ) + .
18
Using (7.76) and (7.77), equation (7.75) becomes,
E=

4
5  5 iv
h5

h f (x0 ) +  f iv (),
15 18
90

(7.78)

## where x0 < < x2 .

This is the error in the interval [x0 , x2 ].
The total error in composite formula is
h5 iv
{f (x0 ) + f iv (x2 ) + + f iv (xn2 )}
90
h5 n iv
=
f ()
90 2
nh5 iv
=
f (),
180
(where f iv () is the maximum among f iv (x0 ), f iv (x2 ), . . . , f iv (xn2 ))
(b a) 4 iv
h f ().
(7.79)
=
180

E=

## Geometrical interpretation of Simpsons 1/3 rule

In Simpsons 1/3 rule, the curve y = f (x) is replaced by the second degree parabola passing through the points A(x0 , y0 ), B(x1 , y1 ) and C(x2 , y2 ). Therefore, the area bounded
by the curve y = f (x), the ordinates x = x0 , x = x2 and the x-axis is approximated to
the area bounded by the parabola ABC, the straight lines x = x0 , x = x2 and x-axis,
i.e., the area of the shaded region ABCDEA.

441

y
6

parabola
C
y = f (x)

A
B

E
O

x0

x1

D x

x2

## Figure 7.3: Geometrical interpretation of Simpsons 1/3 rule.

83
Example 7.11.1 Evaluate 0 (2x x2 ) dx, taking 6 intervals, by (i) Trapezoidal
rule, and (ii) Simpsons 1/3 rule.
Solution. Here n = 6, a = 0, b = 3, y = f (x) = 2x x2 .
30
ba
=
= 0.5.
So, h =
n
6
The tabulated values of x and y are shown below.
x0
xi : 0.0
yi : 0.0
y0

x1
0.5
0.75
y1

x2
1.0
1.0
y2

x3
1.5
0.75
y3

x4 x5 x6
2.0 2.5 3.0
0.0 -1.25 -3.0
y4 y5
y6

## (i) By Trapezoidal rule:

83

h
[y0 + 2(y1 + y2 + y3 + y4 + y5 ) + y6 ]
2
0.5
[0 + 2(0.75 + 1.0 + 0.75 + 0 1.25) 3.0] = 0.125.
=
2
(ii) By Simpsons rule:
83
h
2
0 (2x x ) dx = 3 [y0 + 4(y1 + y3 + y5 ) + 2(y2 + y4 ) + y6 ]
0.5
[0 + 4(0.75 + 0.75 1.25) + 2(1.0 + 0.0) 3.0]
=
3
0.5
[0 + 1 + 2 3] = 0.
=
3
0

(2x x2 ) dx =

## 442 Numerical Analysis

Alternative deduction of Simpsons 1/3 rule
This rule can also be deduced by applying MVT of dierential and of integral calculus.
a+b
ba
a+b ba
+
z = p + qz, p =
,q =
.
Let f C 4 [a, b] and x =
2
2
2
2
Then when x = a, b then z = 1, 1.
Therefore,
%

f (x)dx = q

I=
%

f (p + qz)dz
1

=q

% 1
% 1
g(z)dz +
g(z)dz = q
[g(z) + g(z)]dz

1
% 0

=q
%

1
1

(z)dz,

=q

(7.80)

## where (z) = g(z) + g(z).


Note that (0) = 2g(0) = 2f (p) = 2f ( a+b
2 ), (1) = g(1)+g(1) = f (a)+f (b), (0) =
0.
%
%
1

To prove
0

c.
%

(z)d(z + c) =
0

1+c %

= y(y c)

1+c

(z)dz =
0

## (z)dz = (1 + c)(1) c(0)

c
1+c

(y c)dy

[where z + c = y]

y (y c)dy

= (1 + c)(1) c(0)

(z + c) (z)d(z + c)

= (1 + c)(1) c(0)

(z + c) (z)dz.

(7.81)

## Now, integrating (7.80) thrice

%

(z + c) (z)dz
0
0


1 % 1  z 2
 z 2

+ cz + c1 (z) +
+ cz + c1  (z)dz
= (1 + c)(1) c(0)
2
2
0
0
(z)dz = (1 + c)(1) c(0)

## Differentiation and Integration

443


1
+ c + c1  (1) + c1  (0)
= (1 + c)(1) c(0)
2
 z 3
1 % 1  z 3


z2
z2

+ c + c1 z + c2 (z)
+ c + c1 z + c2  (z)dz
+
6
2
6
2
0
0


1 c
1
+ c + c1  (1) +
+ + c1 + c2  (1)
= (1 + c)(1) c(0)
2
6 2
% 1 3

2
z
z
+ c + c1 z + c2  (z)dz,
c2  (0)
(7.82)
6
2
0
where c1 , c2 , c3 are arbitrary constants and they are chosen in such a way that  (1),  (1)
and  (0) vanish. Thus
1
+ c + c1 = 0,
2

1 c
+ + c1 + c2 = 0,
6 2

and

c2 = 0.

2
1
The solution of these equations is c2 = 0, c1 = , c = .
6
3
Hence
% 1 3

1
2
z 2 z  
z

+
(z)dz
I = q (1) + (0)
3
3
6
3
6
0

1
 4  a + b  h % 1
(z 3 2z 2 + z) (z)dz
=h
f (a) + f (b) + f

3
3
2
6 0


ba
=h
as q =
2
a + b

h
=
f (a) + 4f
+ f (b) + E
3
2
where

%
%
h 1
h 1
2 
z(z 1) (z)dz =
z(z 1)2 [g  (z) g  (z)]dz
E=
6 0
6 0
%
h 1
z(z 1)2 .[2zg iv ()]dz,
z < < z
=
6 0
[by Lagranges MVT]
% 1
h iv
z 2 (z 1)2 dz
[by MVT of integral calculus]
= g (1 )
3
0
h
1
h
0 < 1 < 1.
= g iv (1 ). = g iv (1 ),
3
30
90

Therefore,
h5
E = f iv (2 ).
90

## 444 Numerical Analysis

Hence,

%
a

a + b
h
h5
f (a) + 4f
f (x)dx =
+ f (b) f iv (2 ).
3
2
90

Here, the rst term is the value of the integration obtained from the Simpsons 1/3
rule and the second term is its error.
Algorithm 7.4 (Simpsons 1/3).
8b
a f (x) dx using Simpsons 1/3 rule.

## Algorithm Simpson One Third

Input function f (x);
Read a, b, n; //the lower and upper limits and number of subintervals.//
Compute h = (b a)/n;
Set sum = [f (a) f (a + nh)];
for i = 1 to n 1 step 2 do
Compute sum = sum + 4 f (a + ih) + 2 f (a + (i + 1)h);
endfor;
Compute result = sum h/3;
Print result;
end Simpson One Third.
Program 7.4
.
/* Program Simpsons 1/3
Program to find the value of integration of a function
f(x) using Simpsons 1/3 rule. Here we assume that f(x)=x^3.*/
#include<stdio.h>
void main()
{
float f(float);
float a,b,h,sum;
int i,n;
printf("\nEnter the values of a, b ");
scanf("%f %f",&a,&b);
printf("Enter the value of subintervals n ");
scanf("%d",&n);
if(n%2!=0) {
printf("Number of subdivision should be even");
exit(0);
}
h=(b-a)/n;
sum=f(a)-f(a+n*h);

## Differentiation and Integration

445

for(i=1;i<=n-1;i+=2)
sum+=4*f(a+i*h)+2*f(a+(i+1)*h);
sum*=h/3.;
printf("Value of the integration is %f ",sum);
} /* main */
/* definition of the function f(x) */
float f(float x)
{
return(x*x*x);
}
A sample of input/output:
Enter the values of a, b 0 1
Enter the value of subintervals n 100
Value of the integration is 0.250000
7.11.3

## Simpsons 3/8 rule can be obtained by substituting n =

dierences higher than the third order do not exist here.

% x3
% b
3
3
f (x)dx =
f (x)dx = 3h y0 + y0 + 2 y0 +
2
4
x0
a

3
3
= 3h y0 + (y1 y0 ) + (y2 2y1 + y0 ) +
2
4
3h
[y0 + 3y1 + 3y2 + y3 ].
=
8

## 3 in (7.66). Note that the

1 3
y0
8

1
(y3 3y2 + 3y1 y0 )
8
(7.83)

## This formula is known as Simpsons 3/8 rule.

Now, the interval [a, b] is divided into n (divisible by 3) equal subintervals by the
points x0 , x1 , . . . , xn and the formula is applied to each of the intervals.
Then
% x3
% x6
% xn
% xn
f (x)dx =
f (x)dx +
f (x)dx + +
f (x)dx
x0

x0

x3

xn3

3h
[(y0 + 3y1 + 3y2 + y3 ) + (y3 + 3y4 + 3y5 + y6 )
=
8
+ + (yn3 + 3yn2 + 3yn1 + yn )]
3h
[y0 + 3(y1 + y2 + y4 + y5 + y7 + y8 + + yn2 + yn1 )
=
8
(7.84)
+2(y3 + y6 + y9 + + yn3 ) + yn ].

## 446 Numerical Analysis

This formula is known as Simpsons 3/8 composite rule.
Note 7.11.2 This method is not so accurate as Simpsons 1/3 rule. The error in this
3
formula is h5 f iv (), x0 < < x3 .
80
7.11.4

Booles rule

## Substituting n = 4 in (7.66). The equation (7.66) reduces to

% b
5 2
2 3
7 4
f (x)dx = 4h y0 + 2y0 + y0 + y0 + y0
3
3
90
a
5
2
= 4h[y0 + 2(y1 y0 ) + (y2 2y1 + y0 ) + (y3 3y2 + 3y1 y0 )
3
3
7
+ (y4 4y3 + 6y2 4y1 + y0 )]
90
2h
[7y4 + 32y3 + 12y2 + 32y1 + 7y0 ].
(7.85)
=
45
This rule is known as Booles rule.
8h7 vi
f (), a < < b.
It can be shown that the error of this formula is
945
7.11.5

Weddles rule

## To nd Weddles rule, substituting n = 6 in (7.66). Then

% b
f (x)dx
a

9 2
41 4
11 5
41 6
3
y0
= 6h y0 + 3y0 + y0 + 4 y0 + y0 + y0 +
2
20
20
840

h 6
9
41
11
1
y0 .
= 6h y0 + 3y0 + 2 y0 +43 y0 + 4 y0 + 5 y0 + 6 y0
2
20
20
20
140
h 6
y0 .
If the sixth order dierence is very small, then we may neglect the last term
140
But, this rejection increases a negligible amount of error, though, it simplies the integration formula. Then the above equation becomes
% x6
f (x)dx
x0

3h
[20y0 + 60y0 + 902 y0 + 803 y0 + 414 y0 + 115 y0 + 6 y0 ]
=
10
3h
[y0 + 5y1 + y2 + 6y3 + y4 + 5y5 + y6 ].
=
10
This formula is known as Weddles rule for numerical integration.

(7.86)

447

## Composite Weddles rule

In this rule, interval [a, b] is divided into n (divisible by 6) subintervals by the points
x0 , x1 , . . . , xn . Then
% xn
% x6
% x12
% xn
f (x)dx =
f (x)dx +
f (x)dx + +
f (x)dx
x0

x0

x6

xn6

3h
[y0 + 5y1 + y2 + 6y3 + y4 + 5y5 + y6 ]
=
10
3h
+ [y6 + 5y7 + y8 + 6y9 + y10 + 5y11 + y12 ] +
10
3h
+ [yn6 + 5yn5 + yn4 + 6yn3 + yn2 + 5yn1 + yn ]
10
3h
=
[y0 + 5(y1 + y5 + y7 + y11 + + yn5 + yn1 )
10
+(y2 + y4 + y8 + y10 + + yn4 + yn2 )
+6(y3 + y9 + y15 + + yn3 ) + 2(y6 + y12 + + yn6 )].
(7.87)
The above formula is known as Weddles composite rule.
By the technique used in trapezoidal and Simpsons 1/3 rules one can prove that the
h7 vi
f (), x0 < < x6 .
error in Weddles rule is
140
Degree of Precision
The degree of precision of a quadrature formula is a positive integer n such that the
error is zero for all polynomials of degree i n, but it is non-zero for some polynomials
of degree n + 1.
The degree of precision of some quadrature formulae are given in Table 7.2.
Table 7.2: Degree of precision of some quadrature formulae.
Method
Degree of precision
Trapezoidal
1
Simpsons 1/3
3
Simpsons 3/8
3
Booles
5
Weddles
5

## 448 Numerical Analysis

Comparison of Simpsons 1/3 and Weddles rules
The Weddles rule gives more accurate result than Simpsons 1/3 rule. But, Weddles
rule has a major disadvantage that it requires the number of subdivisions (n) as a
multiple of six. In many cases, the value of h = ba
n (n is multiple of six) is not nite
in decimal representation. For these reasons, the values of x0 , x1 , . . . , xn can not be
determined accurately and hence the values of y i.e., y0 , y1 , . . . , yn become inaccurate.
In Simpsons 1/3 rule, n, the number of subdivisions is even, so one can take n as 10,
20 etc. and hence h is nite in decimal representation. Thus the values of x0 , x1 , . . . , xn
and y0 , y1 , . . . , yn can be computed correctly.
However, Weddles rule should be used when Simpsons 1/3 rule does not give the
desired accuracy.

7.12

## Integration Based on Lagranges Interpolation

Let the function y = f (x) be known at the (n + 1) points x0 , x1 , . . . , xn of [a, b], these
points need not be equispaced.
The Lagranges interpolation polynomial is
(x) =

n

i=0

w(x)
yi
(x xi )w (xi )

(7.88)

where w(x) = (x x0 ) (x xn )
and (xi ) = yi , i = 0, 1, 2, . . . , n.
If the function f (x) is replaced by the polynomial (x) then
%

f (x)dx 

(x)dx =

n %

i=0

w(x)
yi dx.
(x xi )w (xi )

(7.89)

## The above equation can be written as

%

f (x)dx 

where Ci =
a

n


Ci yi ,

(7.90)

i=0
b

w(x)
dx,
(x xi )w (xi )

i = 0, 1, 2, . . . , n.

(7.91)

It may be noted that the coecients Ci are independent of the choice of the function
f (x) for a given set of points.

7.13

449

## Let the interpolation points x0 , x1 , . . . , xn be equispaced, i.e., xi = x0 +ih, i = 1, 2, . . . , n.

Also, let x0 = a, xn = b, h = (ba)/n and yi = f (xi ), i = 0, 1, 2, . . . , n. Then the denite
% b
f (x)dx can be determined on replacing f (x) by Lagranges interpolation
integral
a

% b
n

f (x)dx 
Ci yi ,
a

(7.92)

i=0

## where Ci are some constant coecients.

Now, the explicit expressions for Ci s are evaluated in the following.
The Lagranges interpolation polynomial is
(x) =

n


Li (x)yi ,

(7.93)

i=0

where
Li (x) =

## (x x0 )(x x1 ) (x xi1 )(x xi+1 ) (x xn )

.
(xi x0 )(xi x1 ) (xi xi1 )(xi xi+1 ) (xi xn )

(7.94)

## Introducing x = x0 + sh. Then x xi = (s i)h and xi xj = (i j)h. Therefore,

sh(s 1)h (s i 1)h(s i + 1)h (s n)h
ih(i 1)h (i i 1)h(i i + 1)h (i n)h
(1)ni s(s 1)(s 2) (s n)
.
=
i!(n i)!
(s i)

Li (x) =

(7.95)

## Then (7.92) becomes

% xn
n

f (x)dx 
Ci yi
%

x0

i=0

n
n


(1)ni s(s 1)(s 2) (s n)
yi dx =
or,
Ci yi
(s i)
x0 i=0 i!(n i)!
i=0

n % xn
n


(1)ni s(s1)(s2) (sn)
dx yi =
or,
Ci yi .
(s i)
x0 i!(n i)!
xn

i=0

(7.96)

i=0

## Now, comparing both sides to nd the expression for Ci in the form

% xn
(1)ni s(s 1)(s 2) (s n)
Ci =
dx
(s i)
x0 i!(n i)!
%
(1)ni h n s(s 1)(s 2) (s n)
ds,
=
i!(n i)! 0
(s i)

(7.97)

## 450 Numerical Analysis

i = 0, 1, 2, . . . , n and x = x0 + sh.
ba
Since h =
, substituting
n
Ci = (b a)Hi ,
where
1 (1)ni
Hi =
n i!(n i)!

n
0

(7.98)

s(s 1)(s 2) (s n)
ds, i = 0, 1, 2, . . . , n.
(s i)

## These coecients Hi are called Cotes coecients.

Then the integration formula (7.92) becomes
% b
n

f (x)dx  (b a)
H i yi ,
a

(7.99)

(7.100)

i=0

## where Hi s are given by (7.99).

Note 7.13.1 The cotes coecients Hi s do not depend on the function f (x).
7.13.1 Some results on Cotes coecients
n

Ci = (b a).
(i)
i=0

## By the property of Lagrangian functions,

n

i=0

That is,

% b
n
a i=0

Again,
% b
n
a i=0

w(x)
=1
(x xi )w (xi )

w(x)
dx =
(x xi )w (xi )


w(x)
dx =

(x xi )w (xi )
n

h(1)ni

i=0 0
n


Ci .

dx = (b a).

(7.101)

s(s 1)(s 2) (s n)
ds
i!(n i)!(s i)
(7.102)

i=0

n

i=0

Ci = b a.

(7.103)

(ii)

n


451

Hi = 1.

i=0

## From the relation (7.98),

Ci = (b a)Hi
or,

n


Ci = (b a)

i=0

n


Hi

i=0
n


or, (b a) = (b a)

Hi . [using (7.103)]

i=0

Hence,
n


Hi = 1.

(7.104)

i=0

## That is, sum of cotes coecients is one.

(iii) Ci = Cni .
From the denition of Ci , one can nd
Cni =

(1)i h
(n i)!i!

s(s 1)(s 2) (s n)
ds.
s (n i)

Substituting t = n s, we obtain
Cni

%
(1)i h(1)n 0 t(t 1)(t 2) (t n)
dt
=
i!(n i)!
ti
n
%
(1)ni h n s(s 1)(s 2) (s n)
dt = Ci .
=
i!(n i)! 0
si

Hence,
Ci = Cni .

(7.105)

(iv) Hi = Hni .
Multiplying (7.105) by (b a) and hence obtain
Hi = Hni .

(7.106)

## 452 Numerical Analysis

7.13.2

Trapezoidal rule
Substituting n = 1 in (7.100), we get
%

f (x)dx = (b a)

1


Hi yi = (b a)(H0 y0 + H1 y1 ).

i=0

## Now H0 and H1 are obtained from (7.99) by substituting i = 0 and 1. Therefore,

% 1
% 1
1
1
s(s 1)
ds = and H1 =
sds = .
H0 =
s
2
2
0
0
Here, h = (b a)/n = b a for n = 1.
% b
(b a)
h
(y0 + y1 ) = (y0 + y1 ).
f (x)dx =
Hence,
2
2
a
Simpsons 1/3 rule
%
1 1 2
1
(s 1)(s 2)ds =
For n = 2, H0 = .
2 2 0
6
%
%
1 2
2
1 1 2
1
H2 = .
H1 =
s(s 2)ds = ,
s(s 1)ds = .
2 0
3
2 2 0
6
In this case h = (b a)/2.
Hence equation (7.100) gives the following formula.
%

f (x)dx = (b a)

2


Hi yi = (b a)(H0 y0 + H1 y1 + H2 y2 )

i=0

h
(y0 + 4y1 + y2 ).
3

Weddles rule
To deduce the Weddles rule, n = 6 is substituted in (7.100).
%
a

f (x)dx = (b a)

6


H i yi

i=0

= 6h(H0 y0 + H1 y1 + H2 y2 + H3 y3 + H4 y4 + H5 y5 + H6 y6 )
= 6h[H0 (y0 + y6 ) + H1 (y1 + y5 ) + H2 (y2 + y4 ) + H3 y3 ].
To nd the values of Hi s one may use the result Hi = Hni . Also the value of H3
can be obtained by the formula

## Differentiation and Integration

H3 = 1 (H0 + H1 + H2 + H4 + H5 + H6 ) = 1 2(H0 + H1 + H2 ).
%
41
1 1 6 s(s 1)(s 2) (s 6)
ds =
.
Now, H0 =
6 6! 0
s
840
216
27
272
, H2 =
, H3 =
.
Similarly, H1 =
840
840
840
Hence,
% b
h
[41y0 +216y1 +27y2 +272y3 +27y4 +216y5 +41y6 ].
f (x)dx =
140
a

453

(7.107)

## Again, we know that 6 y0 = y0 6y1 + 15y2 20y3 + 15y4 6y5 + y6 ,

h
h
[y0 6y1 + 15y2 20y3 + 15y4 6y5 + y6 ] 140
6 y0 = 0.
i.e., 140
Adding left hand side of above identity (as it is zero) to the right hand side of (7.107).
After simplication the equation (7.107) nally reduces to
% b
3h
h 6
[y0 + 5y1 + y2 + 6y3 + y4 + 5y5 + y6 ]
y0 .
f (x)dx =
10
140
a
The rst term is the well known Weddles rule and the last term is the error in
Table 7.3: Weights of Newton-Cotes integration rule for dierent n.
n
1
2
3
4
5
6

7.14

1
2
1
3
3
8
14
45
95
288
41
140

1
2
4
3
9
8
64
45
375
288
216
140

1
3
9
8
24
45
250
288
27
140

3
8
64
45
250
288
272
140

14
45
375
288
27
140

95
288
216 41
140 140

## Newton-Cotes Formulae (Open Type)

All the formulae based on Newton-Cotes formula developed in Section 7.13 are of closed
type, i.e., they use the function values at the end points a, b of the interval [a, b] of
integration. Here, some formulae are introduced those take the function values at equispaced intermediate points, but, not at the end points. These formulae may be used

## 454 Numerical Analysis

when the function has singularity(s) at the end points or the values of the function are
unknown at the endpoints. Also, these methods are useful to solve ordinary dierential equations numerically when the function values at the end points are not available.
These formulae are sometimes known as the Steensen formulae.
(i) Mid-point formula
%

x1

x0

1 3 
h f (), x0 x1 .
24

(7.108)

## (ii) Two-point formula

%

x3

f (x)dx =
x0

3h
3h3 
[f (x1 ) + f (x2 )] +
f (), x0 x3 .
2
4

(7.109)

## (iii) Three-point formula

%

x4

f (x)dx =
x0

4h
14h5 iv
[2f (x1 ) f (x2 ) + 2f (x3 )] +
f (),
3
45
x0 x4 .

(7.110)

## (iv) Four-point formula

%

x5

f (x)dx =
x0

5h
[11f (x1 ) + f (x2 ) + f (x3 ) + 11f (x4 )]
24
+

95h5 iv
f (), x0 x5 .
144

(7.111)

## These formulae may be obtained by integrating Lagranges interpolating polynomial

for the data points (xi , yi ), i = 1, 2, . . . , (n 1) between the given limits.

## Methods Based on Undetermined Coecients

In the Newton-Cotes method all the nodes xi , i = 0, 1, 2, . . . , n are known and equispaced. Also, the formulae obtained from Newton-Cotes method are exact for the
polynomials of degree up to n. When the nodes xi , i = 0, 1, 2, . . . , n are unknown then
one can devise some methods which give exact result for the polynomials of degree up
to 2n 1. These methods are called Gaussian quadrature methods.

7.15

455

## The Gaussian quadrature is of the form

% b
n

(x)f (x)dx =
wi f (xi ),
a

(7.112)

i=1

where xi and wi are respectively called nodes and weights and (x) is called the weight
function. Depending on the weight function dierent quadrature formula can be obtained.
The fundamental theorem of Gaussian quadrature states that the optimal nodes of the
m-point Gaussian quadrature formula are precisely the zeros of the orthogonal polynomial for the same interval and weight function. Gaussian quadrature is optimal because
it ts all polynomial up to degree 2m exactly.
To determine the weights corresponding to the Gaussian nodes xi , compute a Lagranges interpolating polynomial for f (x) by assuming
(x) =

m


(x xj ).

(7.113)

j=1

Then
m


(xj ) =

(xj xi ).

(7.114)

i=1
i=j

## Then Lagranges interpolating polynomial through m points is

(x) =

m

j=1

(x)
f (xj )
(x xj )  (xj )

(7.115)

for arbitrary points x. Now, determine a set of points xj and wj such that for a weight
function (x) the following relation is valid.
% b
% b
m
(x)(x)
f (xj )dx
(x)(x)dx =
(x xj )  (xj )
a
a
j=1

m


wj f (xj ),

(7.116)

j=1

1
wj = 
(xj )

%
a

(x)(x)
dx.
x xj

(7.117)

## 456 Numerical Analysis

The weights wj are sometimes called the Christofell numbers.
It can be shown that the error is given by
%
f (2n) () b
E=
(x)[(x)]2 dx.
(2n)! a

(7.118)

Any nite interval [a, b] can be transferred to the interval [1, 1] using linear transformation
x=

b+a
ba
t+
= qt + p.
2
2

(7.119)

Then,
%

f (x) dx =

f (qt + p) q dt.

(7.120)

Thus to study the Gaussian quadrature, we consider the integral in the form
% 1
n

(x)f (x)dx =
wi f (xi ) + E.
(7.121)
1

7.15.1

i=1

## Here (x) is taken as 1 and so the formula (7.121) reduces to

% 1
n

f (x)dx =
wi f (xi ).
1

(7.122)

i=1

It may be noted that wi and xi are 2n parameters and therefore the weights and
nodes can be determined such that the formula is exact when f (x) is a polynomial of
degree not exceeding 2n 1.
Let
f (x) = c0 + c1 x + c2 x2 + + c2n1 x2n1 .

(7.123)

Therefore,
%

f (x)dx =
1

## [c0 + c1 x + c2 x2 + + c2n1 x2n1 ]dx

2
2
= 2c0 + c2 + c4 + .
3
5
When x = xi , equation (7.123) becomes
f (xi ) = c0 + c1 xi + c2 x2i + c3 x3i + + c2n1 x2n1
.
i

(7.124)

457

%

## f (x)dx = w1 [c0 + c1 x1 + c2 x21 + + c2n1 x2n1

]
1

]
+w2 [c0 + c1 x2 + c2 x22 + + c2n1 x2n1
2
+w3 [c0 + c1 x3 + c2 x23 + + c2n1 x2n1
]
3
+
+wn [c0 + c1 xn + c2 x2n + + c2n1 x2n1
]
n
= c0 (w1 + w2 + + wn ) + c1 (w1 x1 + w2 x2 + + wn xn )
+c2 (w1 x21 + w2 x22 + + wn x2n ) +
+c2n1 (w1 x2n1
+ w2 x2n1
+ + wn x2n1
).
n
1
2

(7.125)

Since (7.124) and (7.125) are identical, compare the coecients of ci , and nd 2n
equations as follows:
w 1 + w2 + + w n = 2
w1 x1 + w2 x2 + + wn xn = 0
w1 x21 + w2 x22 + + wn x2n = 23

2n1
+ w2 x2n1
+ + wn x2n1
= 0.
w1 x1
n
2

(7.126)

## Now, equation (7.126) is a set of 2n non-linear equations consisting 2n unknowns wi

and xi , i = 1, 2, . . . , n. Solution of these equations gives the values of wi and xi . Let
wi = wi and xi = xi , i = 1, 2, . . . , n be the solution of (7.126). Then the Gauss-Legendre
formula is nally given by
%

f (x)dx =
1

n


wi f (xi ).

(7.127)

i=1

Unfortunately, determination of general solution of the system (7.126) is very complicated. Thus we concentrate for its particular cases.
Case I. When n = 1, the formula is
%

## f (x)dx = w1 f (x1 ), where w1 = 2 and w1 x1 = 0, i.e., x1 = 0.

Thus for n = 1,
%

f (x)dx = 2f (0).
1

(7.128)

## 458 Numerical Analysis

Case II. When n = 2 then the integral is
% 1
f (x)dx = w1 f (x1 ) + w2 f (x2 ).

(7.129)

w 1 + w2
w1 x1 + w2 x2
w1 x21 + w2 x22
w1 x31 + w2 x32

= 2
= 0
= 23
= 0.

(7.130)

## The above equations can also be obtained by the following way:

The formula (7.129) is exact when f (x) is a polynomial of degree 3. Substituting successively f (x) = 1, x, x2 and x3 in (7.129) and obtain the following system of equations
w1 + w2
w1 x1 + w2 x2
w1 x21 + w2 x22
w1 x31 + w2 x32

=2
=0
= 23
=0

(f (x) = 1)
(f (x) = x)
(7.131)
(f (x) = x2 )
(f (x) = x3 ).

## The solution of these equations is w1 = w2 = 1, x1 = 1/ 3, x2 = 1/ 3. Hence, the

equation (7.129) becomes
% 1

1

## Case III. When n = 3 then the integral becomes

% 1
f (x)dx = w1 f (x1 ) + w2 f (x2 ) + w3 f (x3 ).

(7.132)

## The six unknowns x1 , x2 , x3 and w1 , w2 , w3 are related as

w 1 + w2 + w 3
w1 x1 + w2 x2 + w3 x3
w1 x21 + w2 x22 + w3 x32
w1 x31 + w2 x32 + w3 x32
w1 x41 + w2 x42 + w3 x42
w1 x51 + w2 x52 + w3 x52

= 2
= 0
= 23
= 0
= 25
= 0.

## These equations may also be obtained by substituting f (x) = 1, x, x2 , x3 , x4 , x5 to

the equation (7.132).

459

## Solution of this system of equations is



x1 = 3/5, x2 = 0, x3 = 3/5, w1 = 5/9, w2 = 8/9, w3 = 5/9.
For t